Temporal trends in and factors associated with bleeding complications among patients undergoing percutaneous coronary intervention: a report from the National Cardiovascular Data CathPCI Registry

Abstract

Objectives: The purpose of this study was to examine temporal trends in post-percutaneous coronary intervention (PCI) bleeding among patients with elective PCI, unstable angina (UA)/non-ST-segment elevation myocardial infarction (NSTEMI), and ST-segment elevation myocardial infarction (STEMI).

Background: The impact of bleeding avoidance strategies on post-PCI bleeding rates over time is unknown.

Methods: Using the CathPCI Registry, we examined temporal trends in post-PCI bleeding from 2005 to 2009 among patients with elective PCI (n = 599,524), UA/NSTEMI (n = 836,103), and STEMI (n = 267,632). We quantified the linear time trend in bleeding using 3 sequential logistic regression models: 1) clinical factors; 2) clinical + vascular access strategies (femoral vs. radial, use of closure devices); and 3) clinical, vascular strategies + antithrombotic treatments (anticoagulant ± glycoprotein IIb/IIIa inhibitor [GPI]). Changes in the odds ratio for time trend in bleeding were compared using bootstrapping and converted to risk ratio.

Results: An approximate 20% reduction in post-PCI bleeding was seen (elective PCI: 1.4% to 1.1%; UA/NSTEMI: 2.3% to 1.8; STEMI: 4.9% to 4.5%). Radial approach remained low (<3%), and closure device use increased marginally from 44% to 49%. Bivalirudin use increased (17% to 30%), whereas any heparin + GPI decreased (41% to 28%). There was a significant 6% to 8% per year reduction in annual bleeding risk in UA/NSTEMI and elective PCI, but not in STEMI. Antithrombotic strategies were associated with roughly half of the reduction in annual bleeding risk: change in risk ratio from 7.5% to 4% for elective PCI, and 5.7% to 2.8% for UA/NSTEMI (both p <0.001).

Conclusions: The nearly 20% reduction in post-PCI bleeding over time was largely due to temporal changes in antithrombotic strategies. Further reductions in bleeding complications may be possible as bleeding avoidance strategies evolve, especially in STEMI.

Figure 1. Bleeding Rates Among Elective PCI Population, UA/NSTEMI Population, and STEMI Population Over Time

(A) There was a significant temporal reduction for access-site bleeding, whereas there was no significant reduction in nonaccess site bleeding in the elective percutaneous coronary intervention (PCI) population. (B) There was a significant temporal reduction in access site bleeding for the unstable angina (UA)/non–ST-segment elevation myocardial infarction (NSTEMI) population, whereas there was no significant reduction in nonaccess site bleeding. (C) There was a significant temporal reduction in access site bleeding, and a significant temporal increase in nonaccess site bleeding. CI = confidence interval; OR = odds ratio.

Figure 2. Rate of Radial Artery Catheterization Over Time

The radial approach for catheterization was used infrequently with only a slight increase in use of radial approach over the study period. Abbreviations as in Figure 1.

Figure 3. Vascular Closure Device Utilization Over Time

For each of the 3 groups, there was only a slight temporal increase in the use of vascular closure device utilization over the study period. Abbreviations as in Figure 1.

Figure 4. Changes in Utilization of Different Antithrombotic Strategies Over Time: Elective PCI Population, UA/NSTEMI Population, and STEMI Population

(A) The temporal increase in the use of procedural bivalirudin corresponded with a reduction in the use of heparin + glycoprotein inhibitor (GPI). (B) The temporal increase in the use of procedural bivalirudin corresponded with a reduction in the use of heparin + GPI. The use of bivalirudin in UA/NSTEMI was lower than that seen in the elective PCI population by the end of the study period. (C) The use of heparin + GPI remained the predominant strategy for PCI in the STEMI population. There was a slight increase in the use of bivalirudin toward the end of the study period. Abbreviations as in Figure 1.

Figure 5. Changes in Risk of Annual Bleeding After Accounting for Bleeding Avoidance Strategies

Model 1 demonstrates annual risk reduction in bleeds after adjusting for time and covariates in the CathPCI Bleeding Model. There was minimal attenuation of annual bleeding risk after further adjustment for vascular access strategies (Model 2). Further adjustment for anticoagulation strategies (Model 3) demonstrated a larger attenuation in annual bleeding thereby suggesting changes in anticoagulation strategies are associated with a larger portion of the temporal reduction in post-PCI bleeding events.