Major Bleeding After Lower-Extremity Intervention Exceeds 4%

Debra L Beck

July 01, 2019

Major bleeding is seen in 4.1% of lower-extremity peripheral vascular interventions (PVIs) and is associated with about a 10-fold increased risk for mortality, a large observational analysis suggests.

The study, published in the June 24 issue of JACC: Cardiovascular Interventions, analyzed data from the National Cardiovascular Data Registry (NCDR) PVI registry from 2014 to 2016 and included 18,289 procedures conducted at 76 hospitals in the United States.

Compared with a mortality rate of 0.33% for those with no bleeding, mortality ranged from 3.10% for those with a hemoglobin drop of 4 g/dL or more to a high of 8.80% in those with overt bleeding.

Among those with overt major bleeding, most was access-site bleeding (58.4%), followed by retroperitoneal bleeding (22.6%), gastrointestinal bleeding (3.1%), and genitourinary bleeding (1.4%).

"Of course, the challenge in all of this literature, including our paper, is that it's hard to tell to what extent bleeding is the causative factor in the patient's death or a marker for a high-risk patient," said Adam Salisbury, MD, MSc, Saint Luke's Mid America Heart Institute, Kansas City, Missouri, in an interview with | Medscape Cardiology.

On multivariate analysis, the patient and procedural characteristics associated with bleeding included older age (odds ratio [OR], 1.15 for each decade), female sex (OR, 1.76), previous heart failure (OR, 1.24), low hemoglobin (OR, 1.38), nonelective PCI (OR, 2.50), and critical limb ischemia on presentation (OR, 1.29).

"Many of the risk factors we found that predict bleeding are, of course, the same risk factors that put them at a higher risk for mortality, but what is clear and consistent is that bleeding is associated with bad outcomes and we need to watch these patients very closely because they are at risk for a variety of outcomes, most important, mortality," said Salisbury.

Procedural predictors of bleeding included nonfemoral vascular access (OR, 1.37), use of thrombolytic therapy (OR, 3.44), PVI of the aortoiliac segment (1.89), and multilesion intervention (OR, 1.22).

Bleeding rates varied from 1.2% to 5.9% across institutions, with a median rate of 3.65%. Asked if this might be a function of case mix, Salisbury said probably not.

"It's impossible from observational data to adjust for the impact on outcomes from case mix, but I think we know from the bleeding literature in the coronary space that there's a lot of variability that's not explained just by patient risk factors, which strongly suggest there are practice factors," he said.

To that end, he hopes this work will help inform future investigation, similar to what's been done in the coronary area, that evaluates practice patterns at better-performing centers that can then be introduced to those with higher bleeding rates.

"Stop the Bleeding"

PVI has only recently surpassed bypass surgery as the dominant mode of revascularization for lower extremity PAD. It is also increasingly being performed in the nonhospital, outpatient setting, making it "imperative" that clinicians have a better understanding of the predictors and outcomes of PVI-related bleeding, write Douglas E. Drachman, MD, Massachusetts General Hospital, Boston, and Beau M. Hawkins, MD, University of Oklahoma Health Science Center, Oklahoma City, in an editorial accompanying the publication.

This large, observational analysis by Salisbury and team is a welcome addition to the still-sparse body of literature detailing PCI-related bleeding and best practices, write the editorialists.

For this study, post-PVI bleeding was defined as overt bleeding with a hemoglobin drop of at least 3 g/dL, any hemoglobin drop of at least 4 g/dL, or blood transfusion in patients with preprocedure hemoglobin above 8 d/gL within 72 hours of the procedure.

The bleeding definition used, explained Salisbury, was determined by the data available in the registry. "The NCDR is designed to give us real-world, contemporary practice information, but it's not quite as all-inclusive as a clinical trial where we might look at less severe bleeding," he said.

This "quite stringent" definition of bleeding, write Drachman and Hawkins, may "spur some debate" because it is well known that more bleeding events also affect long-term prognosis in these patients. But, especially given this likely underestimation of clinically significant bleeding, the findings clearly highlight the need for more work in this area.

"For clinicians engaged in the care of patients with lower extremity PAD, this represents an opportunity to establish best practices and improve patient outcomes: it is time to stop the bleeding," they conclude.

"This is really a foundational paper that then leads to the next steps, which are really predicting bleeding risks, identifying best practices, and then applying those practices broadly to reduce variability and care," Salisbury added.

Salisbury and Hawkins reported no relationships relevant to this paper. Drachman reported he is a consultant for Abbott Vascular, Boston Scientific, Broadview Ventures, Cardiovascular Systems, and Corindus Vascular Robotics. He also receives research support from Atrium Medical and Lutonix/CR Bard.

J Am Coll Cardiol Interv. 2019;12:1140-1149. Published online June 24, 2019. Abstract, Editorial


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: