A study using registry data on more than 3.3 million PCI procedures found that patients with major bleeding had a significantly higher inhospital mortality rate compared with patients who did not experience bleeding. The findings were published in the March 13 issue of the Journal of the American Medical Association.
Adnan K. Chhatriwalla, MD, of the biostatistics department at St. Luke’s Mid America Heart Institute, and colleagues wanted to explore the relationships between bleeding risk, bleeding site and mortality. Unlike clinical trials with selected patient populations, they sought to base their analysis on a nationally representative population.
They obtained data from the CathPCI Registry, a resource that includes demographic, procedural and institutional data on cardiac catheterization procedures from more than 1,500 facilities in the U.S. The study period spanned from 2004 to 2011 and included 3,386,688 procedures.
Based on that data, the researchers stratified patients by individual bleeding risk, assessed mortality risk and performed a propensity-matched analysis to calculate the number needed to harm for bleeding-related inhospital mortality. They calculated that the rate of major bleeding was 1.69 percent and the inhospital mortality rate was 0.65 percent. Inhospital mortality was higher in patients with major bleeding compared with no bleeding, at 5.58 percent vs. 0.57 percent.
They determined that 12.1 percent of all inhospital deaths after PCI may be related to bleeding complications. Propensity-matched analyses found that patients with major bleeding had significantly higher inhospital mortality compared with those with no bleeding (5.26 percent vs. 1.87 percent).
The researchers also noted an association between inhospital mortality and major bleeding at all risk levels: low, intermediate and high. The number needed to harm was lowest in patients at high risk of bleeding, at 21, and in patients with nonaccess site bleeding, at 16. Other high-risk subgroups with a lower number needed to harm included the elderly, patients with STEMI and patients with poor renal function.
Chhatriwalla et al wrote that generally physicians “have not embraced” protocols that integrate personalized risk estimates into clinical decision making. “According to these findings, patients in these high-risk subgroups may have the greatest potential for mortality reduction through bleeding avoidance,” they proposed.
Bleeding avoidance strategies may include the use of bivalirudin (Angiomax, The Medicines Company) anticoagulation, arterial closure devices and radial artery access. “These findings raise the possibility that interventions that reduce both access-site and nonaccess-site bleeding, such as bivalirudin, may be associated with greater bleeding-related mortality benefit than would be expected with other bleeding avoidance strategies,” they wrote.
The findings may be affected by unmeasured confounders and may be limited by institution-reported registry data, they acknowledged.