The Safety of a Bivalirudin-Based Approach in Patients undergoing Rotational Atherectomy

§Hitinder S. Gurm, MBBS, ¶Vivek Rajagopal, MD, ¶Deepak L. Bhatt, MD, ¶Stephen G. Ellis, MD,¶A. Michael Lincoff, MD
§Hitinder S. Gurm, MBBS, ¶Vivek Rajagopal, MD, ¶Deepak L. Bhatt, MD, ¶Stephen G. Ellis, MD,¶A. Michael Lincoff, MD
While contemporary percutaneous coronary intervention (PCI) is predominantly a stent-based strategy, other modalities are occasionally used in lieu of or in addition to stents. Rotational atherectomy uses a high-speed rotating bit to pulverize plaque into microparticles and is often used for debulking calcific lesions.1,2 With the widespread use of drug-eluting stents, it is increasingly being used to facilitate stent delivery in calcified vessels.

Previous data indicate that rotational atherectomy is associated with a high risk for periprocedural myonecrosis secondary to the obligate embolization of plaque debris.3 Since secondary platelet aggregation has been implicated in the pathogenesis of this phenomenon, treatment with heparin and platelet glycoprotein (GP) IIb/IIIa receptor inhibitors has been advocated for these patients.4 Recently, bivalirudin with provisional GP IIb/IIIa inhibition has emerged as a viable alternative to heparin and routine GP IIb/IIIa inhibitors in patients undergoing coronary artery interventions.5 It is not clear, however, if this approach is associated with an exaggerated risk of myonecrosis in patients undergoing rotational atherectomy. We accordingly evaluated our experience at a single center with the use of bivalirudin in patients undergoing rotational atherectomy.


Patients undergoing PCI at the Cleveland Clinic are followed in a prospective PCI registry.6 Baseline demographic, clinical, procedural and angiographic characteristics and long-term follow-up data are collected on all patients. Baseline data include patient demographics, acuity of presentation, conventional risk factors, comorbidities, medications, number of diseased vessels, left ventricular ejection fraction, and American College of Cardiology/American Heart Association (ACC/AHA) lesion classification. All patients are followed routinely for enzymatic evidence of infarction. Creatine kinase-MB levels are measured 6–8 hours after the procedure, the morning after the PCI and in the event of suspected ischemia. The study population consisted of 253 patients who underwent rotational atherectomy between January 2001 and December 2004. The physician-operator determined the debulking strategy (device size, number of passes) and adjunct use of platelet GP IIb/IIIa inhibitor. Patients with acute myocardial infarction and those with elevated CK-MB at baseline were excluded from the current analysis. Patients were categorized into those who received a heparin-based versus a bivalirudin-based antithrombotic strategy. Myonecrosis was defined as any rise in CK-MB above the upper limit of normal (ULN), while significant myonecrosis was defined as CK-MB >3 times the ULN. The institutional review board has approved the PCI registry and registry-based analysis.

Statistical analysis. Continuous variables are expressed as mean ± standard deviation, while discrete variables are expressed as frequency counts and percentages. Difference in discrete variables between groups was determined using the chi square test, while the Student’s t-test and the Mann-Whitney test were used for continuous variables. Binary logistic regression was used to calculate the adjusted and unadjusted odds of myonecrosis in association with bivalirudin use. To adjust for the nonrandomized use of bivalirudin therapy in this cohort, a model adjusting for the propensity to receive bivalirudin was developed. The probability of receiving bivalirudin was calculated using a nonparsimonious logistic regression model. The variables included in the models were age, gender, attending physician-operator, date of procedure, noncardiac comorbidities, body weight and height, coronary risk factors, hemodynamic variables, left ventricular ejection fraction, valvular dysfunction, extent of coronary disease, clinical presentation, New York Heart Association Class, ACC/AHA lesion class, and other medications being used at the time of the PCI. Area under the receiver operator characteristics (ROC) curves were used to assess the relative predictive power of the probability scores thus derived. This propensity score was then used to calculate the adjusted odds of myonecrosis and bleeding complication in association with bivalirudin use. Since preprocedural statin use has been demonstrated to reduce the incidence of myonecrosis following rotational atherectomy, we incorporated an interaction term to assess the impact of statin therapy on the safety of a bivalirudin-based approach.7 The survival of the cohort was plotted using the Kaplan-Meier method, and the log rank test was used to determine any differences in outcome. A Cox regression model was used to calculate the adjusted hazard of long-term survival in association with bivalirudin use.


Of the 253 patients who were part of the study, 56 were treated with a bivalirudin-based approach. Table 1 describes baseline clinical, angiographic and procedural data of the cohort dichotomized on the basis of procedural bivalirudin usage. Patients who were treated with bivalirudin were likely to be older, but were less likely to have triple-vessel disease or unstable angina. They were more likely to be treated with drug-eluting stents. Use of bivalirudin increased across the study period, becoming more frequent towards the end of the study. While the use of bivalirudin in the entire PCI population appeared to be associated with the specific operator, no significant association was noted within the subgroup undergoing rotational atherectomy.

Adjunct GP IIb/IIIa inhibition was used in 14 (25%) of patients treated with bivalirudin and 181 (92%) of patients treated with heparin. Among heparin-treated patients, abciximab was used in 150 (76%), eptifibatide in 24 (12.2%) and tirofiban in 7 (3.6%). Abciximab was used in 13 of the bivalirudin patients, while eptifibatide was used in 1 patient. Among the bivalirudin-treated patients, abciximab was used as a bailout strategy for transient slow-flow in 5 cases, while it was used electively in the others. All GP IIb/IIIa use in the heparin-treated patients was planned.

No difference was seen in the presence of other comorbidities (chronic obstructive pulmonary disease, renal dysfunction, peripheral vascular disease) or coronary risk factors (hypertension, diabetes, and smoking) or other angiographic features. The peak activated clotting time (ACT) was higher in patients treated with bivalirudin. Importantly, the use of other preprocedural medications (beta-blockers, statins, angiotensin-converting enzyme inhibitors) did not differ between the two groups.

There was no difference in procedural outcome or complications between the two groups (Table 2). The incidence and severity of postprocedural myonecrosis was similar between the two groups, although there was a trend towards less frequent elevation of CK > 3 times the ULN or 5 times the ULN in patients treated with bivalirudin (Figure 1). Similarly, there was no difference in bleeding complications between the two groups.

Propensity-adjusted model. The propensity score used for discriminating bivalirudin use had an area under the ROC curve of 0.92. In a model adjusting for propensity of being treated with bivalirudin, there was no difference in the odds of myonecrosis or bleeding complications between the two groups (Figure 1). The adjusted odds ratio for mortality and other infrequent complications could not be calculated given their low incidence.

Preprocedural statins and antithrombotic agents. A total of 111 patients were on preprocedural statins in the entire cohort. While there was no difference in the incidence of any myonecrosis among the two treatment strategies (CK ULN 17.9% versus 27.7%; p = 0.45) in patients on statins, the incidence of CK > 3 times the ULN was slightly lower with bivalirudin in the patients pretreated with statins (0% versus 14.5%; p = 0.035). There was, however, no significant interaction between the use of statins and bivalirudin.

Long-term survival. Survival data were available for 204 patients. Among this cohort, there were 28 deaths at a mean follow up of 31 months. The choice of antithrombotic agent did not impact on survival in unadjusted, adjusted or propensity-adjusted models (Figure 2). The significant predictors of survival in this cohort were age, left ventricular ejection fraction, renal dysfunction and diabetes mellitus.


The results of our study indicate that a bivalirudin-based strategy can be safely used in patients undergoing rotational atherectomy. There appears to be no increase in incidence or severity of myonecrosis with this approach compared to patients treated with heparin and GP IIb/IIIa inhibitors.

Rotational atherectomy remains the only plaque-ablative technique that is still currently used in coronary interventions, albeit infrequently. Rotational atherectomy is associated with frequent myonecrosis that is believed to be secondary to embolization of plaque debris down the microvasculature.3,8 Novel observations by Koch and colleagues suggested that the hypoperfusion after rotational atherectomy is common, and was deemed more likely to be related to cavitation or formation of platelet microaggregates.9,10 Further work from the same investigators demonstrated a reduction in the incidence, extent and severity of myocardial hypoperfusion with use of abciximab.11

Even as routine use of rotational atherectomy was displaced by the successful introduction of stents, the antithrombotic strategy for PCI underwent a rapid evolution.12 While the use of heparin only was supplanted by a heparin and routine GP IIb/IIIa inhibitor strategy, there has been a recent shift to increasing use of bivalirudin and provisional GP IIb/IIIa inhibitors. A large body of data supports a reduction in periprocedural bleeding with bivalirudin without a loss in efficacy in patients undergoing PCI using a predominantly stent-based strategy.5,13,14 Given the ease of use, cost factors and the reduction in bleeding, the use of bivalirudin continues to grow.

Our data suggest that such a strategy may possibly work in patients undergoing rotational atherectomy without any evident increase in complications. Given the infrequent use of rotational atherectomy, a randomized trial to directly test the two strategies is unlikely. Although not randomized, our data are important for three reasons. One, they suggest that bivalirudin-based therapy is an option for patients undergoing rotational atherectomy. Secondly, even when we adjusted for covariates and the nonrandomized nature of bivalirudin use, there was no suggestion of an increase in complications. Thirdly, the benefit of bivalirudin was maintained in patients pretreated with statins, given the remarkable reduction in myonecrosis seen with this class of drugs in observational studies.7 Indeed, there was no case of significant myonecrosis among the subset of patients treated with statins and a bivalirudin-based strategy.

Our study did not find a difference in bleeding complications between the two strategies. This may relate to the small number of patients, or may be secondary to the more frequent use of GP IIb/IIIa inhibitors in the bivalirudin-treated patients (compared to historical controls or patients in the REPLACE II trial). Study limitations. Our study is observational in nature and cannot supplant a clinical trial. Although propensity adjustment is an accepted tool for controlling for the nonrandomized nature of drug use, it cannot control for unmeasured variables. It is not known if patients pretreated with clopidogrel and undergoing rotational atherectomy benefit from additional GP IIb/IIIa inhibitor therapy. However, data on preprocedural clopidogrel use were not available for the entire cohort, and hence could not be incorporated into our analysis.


Our findings suggest that a bivalirudin-based strategy with provisional GP IIb/IIIa inhibitors can be safely used in select patients undergoing rotational atherectomy. Further studies are warranted to confirm our findings.






  1. Hoffmann R, Mintz GS, Kent KM, et al. Comparative early and nine-month results of rotational atherectomy, stents, and the combination of both for calcified lesions in large coronary arteries. Am J Cardiol 1998;81:552–557.
  2. Reisman M, Harms V, Whitlow P, et al. Comparison of early and recent results with rotational atherectomy. J Am Coll Cardiol 1997;29:353–357.
  3. Teirstein PS, Warth DC, Haq N, et al. High speed rotational coronary atherectomy for patients with diffuse coronary artery disease. J Am Coll Cardiol 1991;18:1694–1701.
  4. Kereiakes DJ. Preferential benefit of platelet glycoprotein IIb/IIIa receptor blockade: Specific considerations by device and disease state. Am J Cardiol 1998;81:49E–54E.
  5. Lincoff AM, Bittl JA, Harrington RA, et al. Bivalirudin and provisional glycoprotein IIb/IIIa blockade compared with heparin and planned glycoprotein IIb/IIIa blockade during percutaneous coronary intervention: REPLACE-2 randomized trial. JAMA 2003;289:853–863.
  6. Gurm HS, Bhatt DL, Gupta R, et al. Preprocedural white blood cell count and death after percutaneous coronary intervention. Am Heart J 2003;146:692–698.
  7. Gurm HS, Breitbart Y, Vivekanathan D, et al. Preprocedural statin use is associated with a reduced hazard of postprocedural myonecrosis in patients undergoing rotational atherectomy — A propensity-adjusted analysis. Am Heart J 2006;151:1031;e1–e6.
  8. Farb A, Roberts DK, Pichard AD, et al. Coronary artery morphologic features after coronary rotational atherectomy: insights into mechanisms of lumen enlargement and embolization. Am Heart J 1995;129:1058–1067.
  9. Koch KC, Radke PW, Kleinhans E, et al. Mechanisms of myocardial hypoperfusion during rotational atherectomy of de novo coronary artery lesions and stenosed coronary stents: Insights from serial myocardial scintigraphy. J Nucl Cardiol 2002;9:304–311.
  10. Koch KC, Kleinhans E, Klues HG, et al. Quantitative assessment of transient regional ischemia during rotational atherectomy. J Nucl Med 1998;39:402–408.
  11. Koch KC, vom Dahl J, Kleinhans E, et al. Influence of a platelet GP IIb/IIIa receptor antagonist on myocardial hypoperfusion during rotational atherectomy as assessed by myocardial Tc-99m sestamibi scintigraphy. J Am Coll Cardiol 1999;33:998–1004.
  12. Gurm HS, Bhatt DL. Thrombin, an ideal target for pharmacological inhibition: A review of direct thrombin inhibitors. Am Heart J 2005;149:S43–S53.
  13. Gurm HS, Rajagopal V, Fathi R, et al. Effectiveness and safety of bivalirudin during percutaneous coronary intervention in a single medical center. Am J Cardiol 2005;95:716–721.
  14. Lincoff AM, Bittl JA, Kleiman NS, et al. Comparison of bivalirudin versus heparin during percutaneous coronary intervention (the Randomized Evaluation of PCI Linking Angiomax to Reduced Clinical Events [REPLACE]-1 trial). Am J Cardiol 2004;93:1092–1096.