Featured Article

Percutaneous peripheral vascular procedures have increased exponentially over the past 10 years. Anderson et al.¹ reported a 979% growth in peripheral vascular interventions since 1995, with a simultaneous drop in the number of surgical vascular procedures. Despite the sharp rise in peripheral vascular interventions (PVI), complication rates have continued to be significant and, in fact, limb loss has trended upwards.¹

There are many potential predictors of complications in PVI. These predictors potentially include the state of the limb (rest ischemia versus claudication), distal runoff, lesion characteristics (length, occlusion vs. stenosis, location), intraprocedural thrombus, renal insufficiency, age, gender, smoking history, procedure length, number of lesions and vascular beds treated in one setting, and the presence of various comorbidities (diabetes and coronary artery disease). Until recently,^2,3 the choice of the anticoagulant has not been considered as a potential predictor of complications in PVI.

Unfractionated Heparin in PVI
Despite changes in techniques and equipment in the angiographic suite, the use of unfractionated heparin (UFH) has remained the most widely utilized and unchallenged antithrombin during PTA.⁴ We have recently pooled data from 39 studies (15 prospective and 23 retrospective) published in the literature from 1980 to 2000 in order to better define the various complication rates during PVI with UFH as a primary anticoagulant.⁵ All complications within 30 days of PVI were recorded. Only lower extremity and renal procedures were included. Patients who received a primary anticoagulant other than UFH were excluded. Death, stroke, myocardial infarction, major bleed (> 3 gm/dl loss of hemoglobin, retroperitoneal bleed and bleed requiring > 2 units of packed RBC transfusion), renal failure, and limb loss were calculated with the patient number as the denominator. Distal embolization, early occlusion, in-situ thrombosis, vascular complications (pseudoaneurysms and AV fistula) were calculated with the number of vessels angioplastied as the denominator.

The total number of patients reported in this study was 7545 (64.5% males), and the number of vessels angioplastied was 9489. Infrainguinal treatment and total occlusions accounted for 60.9% and 30.8% of all vessels, respectively. There was a high proportion of rest limb ischemia (58.1%), diabetics (38.4%), and coexisting coronary artery disease (41.9%) in these peripheral vascular patients. The overall reported success rate was 89.2% (93.2% for stenosis and 71.7% for occlusions).

In this pooled analysis, complications with UFH were as follows: early occlusion (3.1%), embolization (2.3%), vascular complications (0.7%), major bleed (2.1%), renal failure (1.7%), death (1.6%), limb loss (1.9%), in-situ thrombosis (2.6%), stroke (0.4%), and myocardial infarction (0.7%). The 30-day limb loss and mortality for limb ischemia patients were higher than claudicants (9.1% versus 0.2 % and 2.3% versus 0.1% respectively). This analysis indicates the wide and significant range of complications seen during PVI in patients receiving UFH as a primary anticoagulant and warrants a search for a better and safer replacement.

Bivalirudin in PVI
Bivalirudin, a direct thrombin inhibitor, is a 20-amino acid synthetic peptide that inhibits thrombin directly by binding to both its catalytic site and anion-binding exocite. It produces a dose-dependent prolongation of the ACT with an almost immediate anticoagulant activity following an intravenous injection. The elimination half-life of bivalirudin is approximately 25 minutes in patients with normal renal function.

In contrast to UFH, bivalirudin has a short half-life, provides predictable anticoagulation response, and inhibits free and bound thrombin as well as thrombin-induced platelet activation, aggregation and granule release.⁶ These properties might provide potential benefits over UFH during PTA where thrombin activation is expected to be significant, given the extent of atherosclerotic burden and large vessel size dilated with balloon angioplasty. Furthermore, the short half-life of bivalirudin might also allow early sheath removal, less bleeding complications than UFH, and a more reliable anticoagulation, with no need for frequent ACT measurements during long procedures.^7,8

Several small studies have evaluated bivalirudin in PVI with encouraging findings.^{3, 8-10} Recently, data from the APPROVE (Angiomax Peripheral Procedure Registry of Vascular Events) multi-center registry (25 US centers, 505 patients) was presented by Dr. David Allie, primary investigator, at a recent scientific meeting. In this registry, bivalirudin was utilized as the primary anticoagulant in PVI of renal, iliac and femorals. The REPLACE-2 dose was utilized, and consisted of an 0.75 mg/kg intravenous bolus, followed by 1.75 mg/kg/hour infusion during the length of the procedure. The data from APPROVE appears consistent with previously published bivalirudin data in the periphery where the composite incidence of death, limb loss, unplanned urgent revascularization, and major bleeding appears favorable compared to the historic control UFH. A pooled analysis of published and presented data on bivalirudin in PVI is shown in Table 1.
The bivalirudin group (n= denominator of patients’ number with available data) consisted of 38.2% diabetics (n=808 patients), 36.4% with history of coronary artery disease (n=807), 72.2% suprainguinal vessels (n=832) with 46.4% renals (n=760). Unfortunately, a clear distinction between claudicants and limb ischemia patients was available only on a small number of patients (n=123, excluding renals) and was 4.9%. It is clear, however, that the early experience with bivalirudin seems to be in a low risk population (mostly suprainguinal with a low percentage of limb ischemia patients), which could explain some of the complication differences between it and UFH.

Given the heterogeneous patient populations in both the bivalirudin and UFH groups, and the potential selection bias, a randomized, double-blind trial of bivalirudin versus UFH is warranted.
Early data with bivalirudin is promising, but randomized studies are needed to determine whether this pharmacologic agent is superior to UFH.

The author can be contacted at: shammas@mchsi.com

Dr. Shammas discloses that he is a member of the speakers bureau and has received a research grant from The Medicines Company.