Among patients taking direct oral anticoagulants (DOACs), elective endoscopy procedures carry a risk of bleeding and thromboembolic events similar to that seen in those receiving vitamin K antagonists (VKAs), according to a multicenter, prospective observational study conducted at 12 Spanish academic and community centers.
DOACs have several advantages over VKAs, including more predictable pharmacokinetic profiles and fewer food and drug interactions, but they have not been well studied in the elective endoscopy setting. Some previous studies suggested a lower risk with DOACs than with VKAs, but they were retrospective or based on administrative databases.
It also remains unclear when anticoagulant therapy should be resumed following high-risk procedures. The new study, which was led by Enrique Rodríguez de Santiago of Universidad de Alcalá (Spain) and published in, suggested that early resumption may be safe. “It certainly showed there was an acceptable rate of clinically significant rate of bleeding for patients on anticoagulants, and the thing I appreciated the most was that there was no statistically significant difference in terms of bleeding depending on when you resumed the anticoagulant,” said Robert Jay Sealock, MD, assistant professor of medicine at Baylor College of Medicine in Houston. Dr. Sealock was not involved in the study.
The researchers examined data from 1,623 patients who underwent 1,874 endoscopic procedures. Among these patients, 62.7% were taking VKAs, and 37.3% were taking DOACs; 58.9% were men, and the mean age was 74.2 years. Overall, 75.5% were on anticoagulant therapy for atrial fibrillation.
The most common procedures were colonoscopy (68.3%) and esophagogastroduodenoscopy (27.3%).
Within 30 days, The risk of bleeding was similar between patients taking VKAs (6.2%; 95% confidence interval, 4.8-7.8%) and DOACs (6.7%; 95% CI, 4.9-9%). This was true regardless of intervention and site. Overall, 1.4% of subjects experienced a thromboembolic event (95% CI, 0.9-2.1%), and there was no significant difference between the VKA group (1.3%; 95% CI, 0.8-2.2%) and the DOAC group (1.5%; 95% CI, 0.8-2.8%).
Clinically significant gastrointestinal bleeding occurred in 6.4% of subjects (95% CI, 5.3-7.7%); 2.7% of clinically significant gastrointestinal bleeding events were intraprocedural and 4.1% were delayed. The lowest risk of bleeding occurred with diagnostic endoscopy (1.1%) and biopsy (2.2%). The risk of bleeding for high-risk procedures was 11.5% (95% CI, 9.4-14%).
The overall mortality was 1.4%, with two deaths related to thromboembolic events, both in the DOAC group. The other deaths were considered to be unrelated to the procedure or periprocedural interruption of anticoagulants.
The researchers also examined the timing of anticoagulant resumption. Overall, 59.2% of subjects received bridging therapy, including 85% of the VKA group and 16% of the DOAC group (P < .001). This was not associated with increased endoscopy-related bleeding in either the VKA (3.3% with bridging therapy vs. 6.4% without; P = .14) or the DOAC group (8.3% vs. 6.4%; P = .48).
A total of 747 patients underwent a high-risk procedure, 46.3% of patients resumed anticoagulant therapy within 24 hours of the procedure, and 46.2% between 24 and 48 hours. After inverse probability of treatment weighting adjustment, a delay in anticoagulant resumption was not associated with a reduction in the frequency of postprocedural clinically significant gastrointestinal bleeding.
Still, the research left some questions unanswered. Most of the high-risk procedures were hot (41.8%) or cold snare polypectomies (39.8%). There weren’t enough data in the study to evaluate risk in patients undergoing other high-risk procedures such as balloon dilation for strictures, endoscopic ultrasound with fine-needle aspiration, and sphincterotomy. “That’s one group that we still don’t really have enough data about, particularly those patients who are on DOACs,” said Dr. Sealock.
The study also found a high number of patients on bridging therapy. “It highlighted the fact that we probably use bridging therapy too much in patients undergoing endoscopy,” said Dr. Sealock. He recommended using tools that generate recommendations for bridging therapy and timing for withholding and resuming anticoagulants based on procedure and patient characteristics.
SOURCE: de Santiago ER et al.