Meta-analysis of bleeding complications associated with cardiac rhythm device implantation

Abstract

Background: Many patients receiving cardiac rhythm devices have conditions requiring antiplatelet (AP) and/or anticoagulant (AC) therapy. Current guidelines recommend a heparin-bridging strategy (HBS) for anticoagulated patients with moderate/high risk for thrombosis. Several studies reported lower bleeding risk with continued oral anticoagulation rather than HBS. The best strategy for perioperative management of patients on AP therapy is less clear. The present study was designed as a meta-analysis of device implantation-associated bleeding complications using different AC/AP therapies.

Methods and results: PubMed and Cochrane Database searches identified articles based on design, outcomes, and available data. Device recipients were grouped as follows: no therapy, aspirin only, AC held, AC continued, dual AP, and HBS. The primary outcome was defined as a bleeding complication including hematoma, transfusion, or prolonged hospital stay. Thirteen articles were identified for analysis including 5978 patients. The combined incidence of bleeding complications was 274 of 5978 (4.6%), ranging from 2.2% (no therapy) to 14.6% (HBS). The estimated odds of bleeding were increased by 8.3 (95% CI, 5.5-12.9) times in the HBS group, 5.0 (95% CI, 3.0-8.3) for dual AP therapy, 1.7 (95% CI, 1.0-3.1) for AC held, 1.6 (95% CI, 0.9-2.6) for AC continued, and 1.5 (95% CI, 0.9-2.3) for aspirin only relative to the no therapy group. HBS significantly increased bleeding events compared with holding or continuing AC. Continuing AC did not increase bleeding events compared with no therapy.

Conclusions: Continuing AC appears safer than HBS for device implantation. Dual AP therapy but not continuing AC carries a significant risk of bleeding.

Figure 1

Unadjusted, pooled rates of bleeding complications. Bleeding event rates were (33/1500, 2.2%) for No Therapy ,(26/1044, 2.5%) for AC held, (34/1200, 2.8%) for AC Continued, (45/1165, 3.9%) for SAPT: Single antiplatelet therapy, (37/392, 9.4%) for DAPT: Dual antiplatelet therapy and (99/677, 14.6%) for HBS: Heparin Bridging Strategy.

Figure 2

Forest plot of partial log odds ratios of bleeding complications for six anticoagulant and antiplatelet strategies. Horizontal edges represent 95% credible intervals. AC: Anticoagulant, SAPT: Single antiplatelet therapy, DAPT: Dual antiplatelet therapy, HBS: Heparin Bridging Strategy.

Figure 3

Forest plot of partial log odds for bleeding complications. Horizontal edges represent 95% credible intervals. Negative log odds represent less likelihood of bleeding complications in the given study whereas positive log odds represent increased likelihood of reporting bleeding complications. The two randomized controlled trials are denoted “RCT”.

Figure 4

Funnel plots for each treatment using marginal estimates for the log odds of bleeding complications. Each point represents a single study. Points were labeled by concatenating the first four letters of the author's last name with the last two digits of the year published. The two randomized controlled trials are labeled “Tolo09” and “Chen11”. The number of points within each plot may differ because studies investigated different subsets of the available treatments. Dashed lines represent the mean log odds. Dotted lines form approximate 95% credible ‘funnels’ for log odds, given that no publication bias or study heterogeneity is present. These funnel plots are indicative of study heterogeneity, but not publication bias.