Abstract
Background
Vitamin K antagonists (VKA) are the most widely used anticoagulants, and bridging is commonly administered during periprocedural VKA interruption. Given the unclear benefits and risks of periprocedural bridging in patients with previous venous thromboembolism, we aimed to assess recurrent venous thromboembolism and bleeding outcomes with and without bridging in this population.
Methods
We performed a systematic review searching the PubMed and EMBASE databases from inception to December 7, 2017 for randomized and non-randomized studies that included adults with previous venous thromboembolism requiring VKA interruption to undergo an elective procedure, and that reported venous thromboembolism or bleeding outcomes. Quality of evidence was graded by consensus.
Results
We included 28 cohort studies (20 being single-arm cohorts) with overall 6915 procedures for analysis. In 27 studies reporting perioperative venous thromboembolism outcomes, the pooled incidence of recurrent venous thromboembolism with bridging was 0.7% (95% confidence interval [CI] 0.4–1.2%) and 0.5% (95% CI 0.3–0.8%) without bridging. Eighteen studies reported major and/or non-major bleeding outcomes. The pooled incidence of any bleeding was 3.9% (95% CI 2.0–7.4%) with bridging and 0.4% (95% CI 0.1–1.7%) without bridging. In bridged patients at high thromboembolic risk, the pooled incidence for venous thromboembolism was 0.8% (95% CI 0.3–2.5) and 7.5% (95% CI 3.1–17.4%) for any bleeding. Quality of available evidence was very low, primarily due to a high risk of bias of included studies.
Conclusions
Periprocedural bridging increases the risk of bleeding compared to VKA interruption without bridging, without a significant difference in periprocedural venous thromboembolism rates.
Keywords: bridging, anticoagulants, periprocedural, venous thromboembolism, bleeding
Introduction
Among >6 million individuals in the US who are on chronic anticoagulation, about 250,000 patients each year need to temporarily interrupt their anticoagulants before an invasive procedure to diminish the risk of excess periprocedural bleeding.1 Despite the rapid adoption of direct oral anticoagulants (DOACs) in recent years, vitamin K antagonists (VKAs) remain the most frequently prescribed anticoagulants in the US and worldwide.2, 3 VKAs must be interrupted several days prior to a procedure to allow for regeneration of vitamin K dependent coagulation factors and subsequent normalization of coagulation.1 Because of the concern of an increased risk of thromboembolism during VKA interruption, periprocedural bridging with short-acting parenteral anticoagulants has been recommended for individuals at high thromboembolic risk.1, 4
A previous systematic review investigated periprocedural bridging in patients on VKAs for any indication and found an increased bleeding risk in bridged compared to non-bridged patients without a difference in thromboembolic risk.5 Similarly, a randomized trial of atrial fibrillation patients showed no difference in thromboembolic outcomes, but a significantly higher incidence of major bleeding with bridging compared with placebo.6 However, less is known about the risks and benefits of bridging in patients anticoagulated for venous thromboembolism. A recent retrospective study found an increased bleeding risk with bridging, but no substantial risk of recurrent venous thromboembolism without bridging, irrespective of estimated thromboembolic risk,7 suggesting that current guidelines fail to identify patients with high enough thromboembolic risk to justify bridging.
To better define risks and benefits of bridging in patients with previous venous thromboembolism requiring VKA interruption to undergo an elective invasive procedure, we performed a systematic review comparing recurrent venous thromboembolism and bleeding outcomes with and without periprocedural bridging.
Methods
We conducted this systematic review according to the protocol registered on PROSPERO (registration number CRD42017074710), and reporting conformed to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement.8
Data Sources and Searches
We performed a systematic literature review of articles in PubMed and EMBASE, from inception to December 7, 2017, without language restrictions (Supplemental Methods 1). We also considered conference proceedings, and screened bibliographies of retrieved articles, the most recent American College of Chest Physicians (ACCP) guidelines, and two previous systematic reviews on periprocedural anticoagulation.1, 5, 9 We searched for unpublished literature in the ClinicalTrials.gov and International Clinical Trials Registry Platform (ICTRP) databases.
Study Selection
We included original articles of randomized trials and observational studies that enrolled adults on long-term VKA therapy for secondary prophylaxis of venous thromboembolism who required anticoagulation interruption to undergo elective invasive procedures. Studies were eligible for inclusion if they assessed venous thromboembolism and/or bleeding outcomes in patients with or without heparin bridging; single-arm studies were also considered. Studies that did not specifically report outcomes for the population of interest or studies including <10 venous thromboembolism patients were excluded. Two physicians (C.B. and I.d.K.) independently screened titles and abstracts for eligibility, and in a second step, full-text articles of references selected in the first screening phase were screened (Figure 1). Disagreement was resolved by discussion. Non-English articles were translated by native speakers.
Figure 1.
Study flow diagram. Studies evaluated for inclusion, adapted from PRISMA Statement Flow Diagram.8 Abbreviations: VKA, vitamin K antagonist.
* Until December 7, 2017
† identified through searching of bibliographies and clinical trial databases
‡ refers to duplicate publication (e.g. conference abstract) or studying the same population as another publication
§ because abstract/publication was not available and results could not be obtained from authors, or study was still ongoing
Data Extraction and Quality Assessment
We extracted pre-specified data on study design, population characteristics, thromboembolic risk, invasive procedures, anticoagulation management strategies, and study sponsors. In the absence of a universally accepted definition,1 bridging was classified as receipt of prophylactic, intermediate, or therapeutic doses of any heparin in the perioperative period. Data on patients who continued VKA perioperatively were not included, because our research question related to patients who required interruption of anticoagulation. The primary outcome was recurrent venous thromboembolism including deep vein thrombosis and pulmonary embolism (as ascertained by the individual studies) within 30 days postoperatively.10 Other follow-up periods were used if 30-day outcomes were not reported, as done previously.11 Secondary outcomes included major bleeding (according to the definition used in the individual studies), and any major and non-major bleeding. Outcome data were extracted by two independent investigators. If outcomes were not specifically reported for the population of interest, we contacted study authors to request the information; 3 authors provided additional data.12–14
Two independent reviewers (C.B. and I.d.K) assessed individual study quality using criteria adapted from the ACCP Antithrombotic Therapy and Prevention of Thrombosis, 9th edition;5, 15 this approach was chosen because it is applicable to single-arm studies. Items to assess quality of observational studies included consecutive enrollment, the existence of a study protocol prior to enrolment, the similarity of the setting and time frame of intervention and control groups, blinded outcome assessment, and loss to follow-up.5, 15 Nonrandomized studies were considered to be at high risk of bias and were downgraded to moderate risk if the study accounted for and reported all of the quality elements. On the other hand, they were upgraded to very high risk if they used a single-arm cohort design,16 or if there were severe or multiple problems with these elements.16
Data Synthesis and Analyses
All studies were qualitatively synthesized and comprehensively presented in tables depicting study characteristics and main findings. Meta-analysis to obtain a summary relative risk estimate was not supported because of the observational design of all included studies with reporting of unadjusted results, the high number of single-arm cohorts, and the large clinical heterogeneity across studies. We calculated pooled incidence of venous thromboembolism, major bleeding, and any bleeding in procedures with and without bridging using a logistic-normal random-effects model.17 One study used propensity score matching and weighting to compare outcomes in patients with and without bridging; this resulted in some non-integer numbers of outcome events in the non-bridging group, which were rounded to the next integer to calculate pooled incidences. In a sensitivity analysis, we rounded in the other direction. We also conducted pre-specified sensitivity analyses excluding conference abstracts or small studies (<100 procedures). Additional (not prespecified) sensitivity analyses were conducted: 1) excluding studies that did not specifically report outcomes for patients with prior venous thromboembolism, but could be included because no study patient experienced an outcome event, and 2) excluding a study18 that used propensity score matching and weighting to equalize the number of patients in the exposure groups. Predefined subgroup analyses were conducted according to baseline thromboembolic risk and heparin dose used (therapeutic vs. prophylactic). All analyses were performed with Stata 14 (Stata, College Station, TX, USA).
Two investigators (C.B. and M.C. F.) rated the overall quality of evidence for thromboembolic and bleeding outcomes using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) methodology.19
Results
Our search yielded 4,349 records after removal of duplicates. We excluded 4,110 records based on title and abstract and then assessed 239 full texts for eligibility, of which 28 studies met our inclusion criteria (Figure 1). Included studies reported results of overall 6,915 procedures, and individual study size varied between 10 and 2670 procedures (Supplemental Table 1).
Of all included studies, 19 were prospective and 9 retrospective cohort studies, and none of them was a randomized trial comparing bridging and VKA interruption without bridging. Study characteristics are shown in Table 1 and Supplemental Table 1. Two records were conference abstracts, and 26 were published full text articles. All but 5 studies enrolled patients with different indications for anticoagulation including atrial fibrillation and mechanical heart valves, and among those we extracted only the specific data on patients with previous venous thromboembolism. Thus, specific information on demographics for this subgroup was mostly missing. Baseline thromboembolic risk was assessed in 21 studies using varying risk classifications schemes. Most studies included patients at any thromboembolic risk (n=15), while 1 included only low risk and 5 only moderate to high risk patients. The decision to administer bridging anticoagulation was at the discretion of the provider in 11 studies and based on assessment of baseline thromboembolic risk in 6 studies; in 5 studies, all patients requiring VKA interruption received bridging. Procedure types were reported in all but 1 study.
Table 1.
Characteristics of Included Studies
| Characteristic | Studies, n (%) | |
|---|---|---|
| Design and Setting | ||
| Study design | Randomized controlled trial | 0* |
| Cohort, prospective | 19 (68) | |
| Cohort, retrospective | 9 (32) | |
| Single-arm studies | 20 (71) † | |
| Year of Publication | Before 2010 | 12 (43) |
| 2010 and later | 16 (57) | |
| Publication Type | Full text article | 26 (93) |
| Conference abstract | 2 (7) | |
| Study sponsors | Industry | 5 (18) |
| Non-profit | 7 (25) | |
| Not reported | 16 (57) | |
| Geographical region | North America | 11 (39) |
| South America | 1 (4) | |
| Europe | 11 (39) | |
| Asia | 1 (4) | |
| Australia | 2 (7) | |
| Not reported | 2 (7) | |
| Number of study sites | Single center | 14 (50) |
| Multicenter | 10 (36) | |
| Not reported | 4 (14) | |
| Treatment setting | Outpatients only | 4 (14) |
| Population | ||
| % of patients with previous venous thromboembolism | 0–19% | 10 (36) |
| 20–99% | 13 (46) | |
| 100% | 5 (18) | |
| Thromboembolic risk of included patients | Any risk | 15 (54) |
| Only low risk | 1 (4) | |
| Only moderate to high risk | 5 (18) | |
| No risk stratification performed | 7 (25) | |
| Reason for VKA interruption | Antiarrhythmic device surgery | 2 (7) |
| Dental procedures | 1 (4) | |
| Only minor surgeries or procedures | 4 (14) | |
| Minor and major surgeries/procedures | 20 (71) | |
| Unclear | 1 (4) | |
| Indication for bridging | At the discretion of the provider | 11 (39) |
| Strictly based on thromboembolic risk | 6 (21) | |
| All patients requiring VKA interruption | 5 (18) | |
| Random assignment | 2 (7) ‡ | |
| Only inpatients | 1 (4) | |
| unclear | 3 (11) | |
Abbreviations: INR, international normalized ratio; VKA, vitamin K antagonist.
three studies were designed as randomized controlled trials comparing bridging to continued VKA or comparing two different bridging strategies. We considered these studies as prospective cohorts for our study because none of these trials compared bridging and VKA interruption without bridging.
among all prospective and retrospective cohort studies
patients were randomly assigned to bridging or continuation of VKA in two studies, but none used random assignment to bridging vs. VKA interruption without bridging
Some studies implemented multiple strategies
Bridging strategies implemented in the different studies varied substantially (Table 2 and Supplemental Table 2). Most of the studies (n=22) defined bridging as administration of short-acting anticoagulants in both the pre- and postprocedural periods, while 2 studies also considered bridging that was administered in the preprocedural and 4 in the postprocedural periods only.
Table 2.
Periprocedural Anticoagulation Management
| Management Strategy* | Studies, n (%) | |
|---|---|---|
| Periprocedural Bridging | Pre- and postprocedural | 22 (79) |
| Only preprocedural | 2 (7) | |
| Only postprocedural | 4 (14) | |
| Not specified | 4 (14) | |
| Type of bridging anticoagulant | LMWH | 28 (100) |
| Unfractionated heparin | 6 (21) | |
| Dosing of bridging anticoagulant | Therapeutic | 18 (64) |
| Sub-therapeutic | 9 (32) | |
| Prophylactic | 16 (57) | |
| Not reported | 1 (4) | |
| Preprocedural Management | ||
| VKA interruption, days | ≥6 | 7 (25) |
| 5 | 19 (68) | |
| ≤4 | 9 (32) | |
| not reported | 3 (11) | |
| Last dose of LMWH, hours | < 12 | 1 (4) |
| 12–23 | 13 (46) | |
| ≥24 | 7 (25) | |
| NA (only postop bridging) | 2 (7) | |
| Not reported | 6 (21) | |
| Postprocedural Management | ||
| Restart of LMWH, hours | 0–23 | 16 (57) |
| ≥24 | 7 (25) | |
| Not specifically reported | 7 (25) | |
| Restart of VKA, hours | Evening of procedure | 10 (36) |
| Day after procedure | 18 (64) | |
| 2 or more days after procedure | 4 (14) | |
| Not reported | 6 (21) | |
| Initial dosing of VKA | Maintenance dose | 9 (32) |
| Loading dose | 3 (11) | |
| Not reported | 16 (57) | |
| Duration of bridging | Prespecified no. of days | 2 (7) |
| Until INR therapeutic | 12 (43) | |
| Until INR therapeutic on 2 consecutive days | 5 (18) | |
| Not reported | 9 (32) | |
Abbreviations: INR, international normalized ratio; LMWH, low molecular weight heparin; VKA, vitamin K antagonist.
Some studies implemented multiple strategies
Overall quality of included studies was low (Supplemental Table 3). Only 8 studies compared outcomes in bridging and non-bridging groups, while the remaining were single-arm studies. All studies reported results that were unadjusted for potential confounding except for Sjögren et al.,18 where propensity score matching was used.
Risk of Venous Thromboembolism
All but one20 study reported venous thromboembolism outcomes in overall 6886 procedures. Over a follow-up duration that varied from the time of postprocedural INR normalization to 180 days in the different studies, incidence of perioperative venous thromboembolism ranged from 0% to 4.4% with bridging (pooled incidence 0.7%, 95% confidence interval [CI] 0.4–1.2%) and from 0% to 2% without bridging (pooled incidence 0.5%, 95% CI 0.3–0.8%; Tables 3 and 4).
Table 3.
Venous Thromboembolism and Bleeding Outcomes
| Bridging | No Bridging | ||||||
|---|---|---|---|---|---|---|---|
| Study | Follow-up Duration | Venous thromboembolism/ Procedures, n (%) | Major Bleeding/ Procedures, n (%) | Any Bleeding ||/ Procedures, n (%) | Venous thromboembolism/ Procedures, n (%) | Major Bleeding/ Procedures, n (%) | Any Bleeding ||/ Procedures, n (%) |
| Ahmed, 201021 | 8 weeks | 0/12 (0) | NA | 1/12 (8.3) | 0/5 (0) | NA | 0/5 (0) |
| Bajkin, 200922* | 30d | 0/19 (0) | NA | not clear | - | - | - |
| Birnie, 201323* | not clear | 0/16 (0) | NA | not clear | - | - | - |
| Breen, 201624 | 30d | 0/29 (0) | not clear | 6/29 (20.7) | - | - | - |
| Burbury, 201125* | 6 weeks | - | - | - | 0/39 (0) | not clear | not clear |
| Clark, 20157 | 30d | 0/555 (0) | 12/555 (2.2) | 15/555 (2.7) | 3/1257 (0.2) | 2/1257 (0.2) | 2/1257 (0.2) |
| Constans, 200742* | 90d | 0/10 (0) | not clear | not clear | - | - | - |
| Douketis, 200520 | until INR 2–3 | NA | 0/21 (0) | 0/21 (0) | - | - | - |
| Dunn, 200726 | 28d | 1/96 (1.0) | not clear | not clear | - | - | - |
| Eisele, 201227* | not clear | 0/69 (0) | NA | not clear | - | - | - |
| Fernandez, 200928* | 30d | 0/24 (0) | not clear | not clear | 0/14 (0) | 0/14 (0) | 0/14 (0) |
| Garcia, 200829 | 30d | 0/22 (0) | not clear | not clear | 2/179 (1.1) | 0/179 (0) | 0/179 (0) |
| Hammerstingl, 200930* | not clear | 0/10 (0) | 0/10 (0) | 0/10 (0) | - | - | - |
| Hwang, 201731* | 30d | 0/34 (0) | not clear | not clear | - | - | - |
| Jaffer, 200532 | 30d | 0/18 (0) | 1/18 (5.6) | 1/18 (5.6) | - | - | - |
| Klamroth, 201012 | 28d | 0/63 (0) | 0/63 (0) † | 0/63 (0) † | - | - | - |
| Majluf-Cruz, 201133 | not clear | 2/152 (1.3) | 4/152 (2.6) | 26/152 (17.1) | - | - | - |
| Malato, 201034 | 30d | 2/45 (4.4) | 1/45 (2.2) | 1/45 (2.2) | - | - | - |
| McBane, 201013 | 90d | 4/342 (1.2) † | 9/342 (2.6) † | 24/342 (7.0) † | 3/152 (2.0) † | 1/152 (0.7) † | 3/152 (2.0) † |
| Pengo, 200914 | 30d | 2/210 (1.0) | 1/210 (0.5) † | 9/210 (4.3) † | - | - | - |
| Saccullo, 201235 | 30d | 2/52 (3.8) | not clear | not clear | - | - | - |
| Santamaria, 201336* | 90d | 0/10 (0) | not clear | not clear | - | - | - |
| Sjögren, 201718 | 30d | 10/1331 (0.8) | NA | 10/1331 (0.8) | 5.5/1331 (0.4) ‡ | NA | 2.1/1331 (0.2) ‡ |
| Skeith, 201237 | 90d | 2/152 (1.3) | 2/152 (1.3) | 6/152 (3.9) | 2/482 (0.4) | 8/482 (1.6) | 16/482 (3.3) |
| Spyropoulos, 200438 | 60d | 0/45 (0) | 0/45 (0) | 2/45 (4.4) | - | - | - |
| Spyropoulos, 200639* | 30d | 0/22 (0) § | not clear | not clear | - | - | - |
| Wilson, 200140 | 90d | 0/26 (0) | 0/26 (0) | 0/26 (0) | - | - | - |
| Wiszniewsky, 201141 | 180d | 0/63 (0) | 0/63 (0) | 13/63 (20.6) | - | - | - |
Abbreviations: NR, not available because outcome was not assessed in study.
these studies did not specifically report venous thromboembolism outcomes by indication for anticoagulation (i.e. patients with previous venous thromboembolism), but were included because the number of outcome events was 0 for all indications
personal communication by the authors
patients not receiving bridging anticoagulation were propensity score matched and weighted to equal the number of patients who received bridging
the number of venous thromboembolic events in subgroup of patients with a previous venous thromboembolism could only be identified in the 22 patients who received unfractionated heparin
includes major bleeding and any other bleeding outcome that was specifically reported for patients with previous venous thromboembolism
Table 4.
Pooled Incidence of Venous Thromboembolism and Bleeding Outcomes
| Any Outcome | Venous Thromboembolism | Major Bleeding | All Bleeding* | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Group | Studies, n | Procedures, n |
Studies, n | Events/ Procedures, n | Pooled incidence, % (95% CI) | Studies, n | Events/ Procedures, n | Pooled incidence, % (95% CI) | Studies, n | Events/ Procedures, n | Pooled incidence, % (95% CI) |
| Main Analysis | |||||||||||
| All studies | 28 | 6915† | 27 | 40.5 / 6886 | 15 | 41 / 3786 | 18 | 137.1/ 6494 | |||
| With bridging | 27 | 3448 | 26 | 25 / 3427 | 0.7 (0.4–1.2) | 13 | 30 / 1702 | 1.8 (1.2–2.5) | 16 | 114 / 3074 | 3.9 (2.0–7.4) |
| Without bridging | 8 | 3459 | 8 | 15.5 / 3459 | 0.5 (0.3–0.8)‡ | 5 | 11 / 2084 | 0.4 (0.1–1.5) | 7 | 23.1 / 3420 | 0.4 (0.1–1.7)§ |
| Sensitivity analyses | |||||||||||
| Excluding studies without specific outcome reporting for population of interest || | 18 | 6648† | 17 | 40.5 / 6619 | 11 | 41 / 3557 | 14 | 137.1/ 6265 | |||
| With bridging | 18 | 3234 | 17 | 25 / 3213 | 0.8 (0.5–1.3) | 11 | 30 / 1666 | 1.8 (1.3–2.6) | 14 | 114 / 3038 | 4.4 (2.3–8.2) |
| Without bridging | 6 | 3406 | 6 | 15.5 / 3406 | 0.5 (0.3–0.8)‡ | 3 | 11 / 1891 | 0.5 (0.1–2.0) | 5 | 23.1 / 3227 | 0.6 (0.1–2.4)¶ |
| Excluding conference abstracts | 26 | 6725† | 25 | 38.5 / 6696 | 13 | 37 / 3620 | 16 | 111.1/ 6328 | |||
| With bridging | 25 | 3272 | 24 | 23 / 3251 | 0.7 (0.4–1.2) | 12 | 26 / 1550 | 1.6 (0.8–3.2) | 15 | 88 / 2922 | 3.4 (1.8–6.6) |
| Without bridging | 7 | 3445 | 7 | 15.5 / 3445 | 0.5 (0.3–0.8)‡ | 4 | 11 / 2070 | 0.4 (0.1–1.6) | 6 | 23.1 / 3406 | 0.4 (0.1–1.8)** |
| Excluding Sjögren, et al. † | 27 | 4245 | 26 | 25 / 4224 | 15 | 41 / 3832 | 17 | 125 / 3878 | |||
| With bridging | 26 | 2117 | 25 | 15 / 2096 | 0.6 (0.3–1.3) | 13 | 30 / 1702 | 1.8 (1.2–2.5) | 15 | 104 / 1743 | 4.9 (2.7–8.7) |
| Without bridging | 7 | 2128 | 7 | 10 / 2128 | 0.5 (0.2–1.2) | 5 | 11 / 2084 | 0.4 (0.1–1.5) | 6 | 21 / 2089 | 0.6 (0.1–2.8) |
| Excluding studies with <100 procedures | 7 | 6173† | 7 | 35.5 / 6165 | 6 | 39 / 3481 | 7 | 113.1/ 6143 | |||
| With bridging | 7 | 2764 | 7 | 20 / 2764 | 0.7 (0.4–1.2) | 5 | 28 / 1411 | 2.0 (1.4–2.9) | 6 | 90 / 2742 | 4.0 (1.8–8.7) |
| Without bridging | 5 | 3401 | 5 | 15.5 / 3401 | 0.5 (0.3–0.8)‡ | 4 | 11 / 2070 | 0.4 (0.1–1.6) | 5 | 23.1 / 3401 | 0.5 (0.1–1.9) |
| Subgroup analyses | |||||||||||
| High thromboembolic risk †† | 9 | 394 | 9 | 3 / 394 | 7 | 45 / 361 | |||||
| With bridging | 9 | 373 | 9 | 3 / 373 | 0.8 (0.3–2.5) | NA | NA | NA | 7 | 44 / 340 | 7.5 (3.1–17.4) |
| Without bridging | 1 | 21 | 1 | 0 / 21 | NA | NA | NA | NA | 1 | 1 / 21 | NA |
| LMWH dose †† | |||||||||||
| Therapeutic dose | 11 | 981 | 10 | 5 / 580 | 0.9 (0.4–2.1) | 4 | 10 / 387 | 2.6 (1.4–4.7) | 6 | 35 / 800 | 3.9 (2.0–7.5) |
| Prophylactic dose | 6 | 362 | 5 | 3 / 208 | 1.4 (0.5–4.4) | 3 | 4 / 188 | 2.1 (0.8–5.5) | 4 | 32 / 342 | 4.4 (0.9–18.3) |
includes major bleeding and any other bleeding outcomes that were specifically reported for patients with previous venous thromboembolism
in the study by Sjögren et al,18 procedures in the non-bridging group were propensity score matched and weighted to equal the number of procedures in the bridging group; therefore, the number of procedures with bridging and without bridging does not add up to the exact number of overall procedures
the non-integer number of outcome events (5.5 venous thromboembolic events) in the study by Sjögren et al was rounded to the next integer for this analysis. When rounding down to the next lower integer, the results were similar (pooled incidence 0.4%, 95%CI 0.3–0.7%)
the non-integer number of outcome events (2.1 events of any bleeding) in the study by Sjögren et al was rounded to the next integer for this analysis. When rounding up to the next higher integer, the results were similar (pooled incidence 0.5%, 95%CI 0.1–1.8%)
some studies did not report results separately for patients with prior venous thromboembolism and could only be included because there were no venous thromboembolism 22, 23, 25, 27, 28, 30, 31, 36, 39, 42 or bleeding28–30, 40 outcomes in the overall population. As this might bias our study sample to studies with a low number of outcome events, we excluded these studies for this subgroup analysis.
the non-integer number of outcome events (2.1 events of any bleeding) in the study by Sjögren et al was rounded to the next integer for this analysis. When rounding up to the next higher integer, the results were similar (pooled incidence 0.7%, 95%CI 0.2–2.4%)
the non-integer number of outcome events (2.1 events of any bleeding) in the study by Sjögren et al was rounded to the next integer for this analysis. When rounding up to the next higher integer, the results were similar (pooled incidence 0.5%, 95%CI 0.1–1.9%)
for these analyses, we included only studies that reported outcomes separately for the specific subgroup of interest
The pooled incidence of venous thromboembolism in bridging and non-bridging groups remained similar in sensitivity analyses (Table 4). Among patients with high thromboembolic risk, we found a similar pooled incidence of perioperative venous thromboembolism events with bridging (pooled incidence 0.8%, 95% CI 0.3–2.5%). Only 1 study7 reported venous thromboembolism outcomes in high risk patients without bridging, and found 0 events in 21 procedures (Table 4). Incidence of venous thromboembolism was similar with therapeutic LMWH doses compared to prophylactic doses (pooled incidence 0.9% [95% CI 0.4–2.1%] and 1.4% [95% CI 0.5–4.4%], respectively), but the number of studies included in this subgroup analysis was limited (Table 4).
Risk of Bleeding
Data on major bleeding or any bleeding were available in 15 and 18 studies, respectively (Tables 3 and 4). Incidence of major bleeding as reported in included studies varied between 0% and 5.6% for patients with bridging (pooled incidence 1.8%, 95% CI 1.2–2.5%), while it was generally lower in patients without bridging, ranging from 0% to 1.6% (pooled incidence 0.4%, 95% CI 0.1–1.5%; Tables 3 and 4). Results were similar in sensitivity analyses (Table 4). Therapeutic LMWH doses resulted in a similar pooled incidence of major bleeding compared to prophylactic doses (2.6% [95% CI 1.4–4.7%] and 2.1% [95% CI 0.8–5.5%]), but the number of studies was low.
Incidence of any major and non-major bleeding differed even more considerably between the bridging and non-bridging groups, with a pooled incidence of 3.9% (95% CI 2.0–7.4%) and 0.4% (95% CI 0.1–1.7%), respectively (Table 3 and 4).
Quality of Overall Evidence
The quality of the overall body of evidence for the association of periprocedural bridging with venous thromboembolism, major bleeding and any bleeding was very low (Supplemental Table 4). All included studies had an observational design with a very high risk of bias. Also, we assumed publication bias to be substantial for all outcomes, because some studies that did not separately report outcomes for our specific population of interest could not be included in our systematic review (Figure 1), and a search in clinical trial registries yielded two studies without published results that potentially met inclusion criteria.
Discussion
This systematic review showed that periprocedural bridging in patients with previous venous thromboembolism considerably increases bleeding risk compared to VKA interruption without bridging, without resulting in differences of venous thromboembolic outcomes. Our results suggest that most venous thromboembolism patients will not benefit from bridging. However, the quality of the available evidence on the risks and benefits of periprocedural bridging in this population is low.
Our results show substantial differences in the definition and management of periprocedural bridging across studies. However, no specific bridging strategy or dosing regimen has proven to be superior.4 To date, no randomized study has shown a clear benefit of periprocedural bridging for patients on long-term anticoagulants, while the bulk of available data suggests an increased bleeding risk with bridging. A recent randomized trial found that foregoing bridging was not associated with an increased risk of arterial thromboembolic events but conferred a significant reduction in major bleeding in patients with atrial fibrillation requiring periprocedural warfarin interruption.6 Another trial comparing bridging with continued anticoagulation in patients undergoing cardiac device surgery similarly found a 3- to 4-fold higher bleeding risk with bridging.23 A systematic review that assessed periprocedural bridging in patients with any indication for anticoagulation concluded that overall and major bleeding was increased in patients with bridging compared to those without, while the risk of thromboembolism did not differ.5 However, the low quality of included studies precluded firm conclusions about risks and benefits of bridging particularly for high risk patients, similar to our study. No previous or ongoing trial investigated bridging in patients with previous venous thromboembolism.
Evaluation of patient- and procedure-related risk factors for thromboembolism and bleeding as well as estimation of clinical consequences of these potential adverse events is needed to determine the optimal perioperative anticoagulation management in patients on chronic VKA therapy.43 An increasing evidence base suggests that VKAs can be safely continued in a number of procedures with low bleeding risk.6, 22, 23, 44 More uncertainty exists about periprocedural management strategies in non-minimal bleeding risk procedures, where anticoagulation interruption is needed. While arterial thromboembolic events in patients with mechanical heart valves or atrial fibrillation are related to a substantial 15–70% risk of disability and death which outweighs mortality risks of major bleeding events 2 to 10-fold,43, 45, 46 the case-fatality of recurrent venous thromboembolism (5–13%) and major bleeding events (8–10%) are more similar.43, 47 Therefore, the number of major bleeding events that would theoretically be acceptable to prevent one thromboembolic event is substantially lower for patients with previous venous thromboembolism compared to patients anticoagulated for prevention of arterial thromboembolic events, so the threshold to provide treatments that increase the risk of major bleeding should be higher for venous thromboembolism patients. Our results suggest that bridging exposes these patients to a markedly higher bleeding risk and should therefore not be used if thromboembolic risk is low or moderate. Based on the available evidence, admittedly of low quality, a potential benefit of bridging has not been established even in patients at high thromboembolic risk.
Periprocedural bridging continues to be overused in patients with previous venous thromboembolism, mainly because of overestimation of thromboembolic risk,7, 48 reflecting the uncertainty about the role of bridging and the assessment of baseline thromboembolic risk. Efforts to standardize anticoagulation management in the periprocedural setting and high quality studies to further clarify the efficacy and safety of bridging in patients at high risk of venous thromboembolism are needed to optimize their care. Although VKAs continue to be the most frequently prescribed oral anticoagulants,2 the role of bridging, if any, will further diminish with the increased use of DOACs.4, 49 The shorter half-life of DOACs, which limits the duration of preprocedural anticoagulation interruption, and their rapid time-to-onset obviate the need of periprocedural bridging.4, 43, 49 Only few observational studies and substudies of randomized trials investigated periprocedural management strategies in patients on DOACs, and did not find a benefit of bridging in this population.50, 51
To our knowledge, this is the first study systematically assessing available literature on the risks and benefits of periprocedural bridging in the specific population of patients with previous venous thromboembolism. Another strength of this study includes the rigorous search for available data, without language restriction and with consideration of unpublished articles.
The results of our systematic review need to be interpreted in the context of several limitations. First, overall quality of included studies was low: all included studies were non-randomized and many of them lacked a comparison group. Second, most studies included a heterogeneous population of patients at high and low thromboembolic risk and procedures at varying degrees of bleeding risk, yielding difficult to interpret results that are likely affected by confounding by indication, as patients at high thromboembolic risk might be more likely to receive bridging compared to non-high risk patients. Third, some studies did not specifically report outcomes for patients with previous venous thromboembolism and could thus not be included in our systematic review (Figure 1), potentially resulting in publication bias. Fourth, follow-up duration differed between studies, and although we calculated pooled incidences of perioperative outcome events, these incidences might differ if follow-up duration in included studies were more homogeneous. Finally, the definition of major bleeding and ascertainment of thromboembolic events was not uniform across studies, which may have limited the reliable estimation of pooled outcome incidences.
In conclusion, our systematic review showed that patients at low and moderate thromboembolic risk do not benefit from periprocedural bridging because of a considerably increased risk of bleeding associated with bridging compared to VKA interruption without bridging, while the risk of periprocedural thromboembolic events was similar with or without bridging.
Supplementary Material
Acknowledgments
We gratefully acknowledge Isabel Elaine Allen for her advice concerning the methodology and statistical analyses of the paper. We acknowledge Yumiko Abe-Jones and Michael Hurley for their help with translating articles, and James D. Douketis for the critical review of the manuscript and his valuable input.
Sources of Funding
This study was supported by the National Heart, Lung, and Blood Institute (Grant R01HL103820 and 1K24HL141354). C. Baumgartner’s work was supported by a grant from the Swiss National Science Foundation (P2BEP3_165409) and a grant from the Gottfried and Julia Bangerter-Rhyner Foundation.
Role of the Sponsor
The sponsor had no role in in the design and conduct of the study, in the collection, management, analysis, or interpretation of the data, or the preparation, review, or approval of the manuscript.
Registration: PROSPERO, registration number CRD42017074710
Footnotes
Disclosures
None.
References
- 1.Douketis JD, Spyropoulos AC, Spencer FA, Mayr M, Jaffer AK, Eckman MH, Dunn AS, Kunz R and American College of Chest P. Perioperative management of antithrombotic therapy: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141:e326S–350S. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Medical Expenditure Panel Survey (MEPS) 2005–2015. Agency for Healthcare Research and Quality (AHRQ), Rockville, MD: ClinCalc DrugStats Database version 18.0 http://clincalc.com/DrugStats/Top300Drugs.aspx, accessed April 11, 2018. [Google Scholar]
- 3.Lippi G, Mattiuzzi C, Cervellin G and Favaloro EJ. Direct oral anticoagulants: analysis of worldwide use and popularity using Google Trends. Annals of translational medicine 2017;5:322. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Doherty JU, Gluckman TJ, Hucker WJ, Januzzi JL, Jr., Ortel TL, Saxonhouse SJ and Spinler SA. 2017 ACC Expert Consensus Decision Pathway for Periprocedural Management of Anticoagulation in Patients With Nonvalvular Atrial Fibrillation: A Report of the American College of Cardiology Clinical Expert Consensus Document Task Force. J Am Coll Cardiol 2017;69:871–898. [DOI] [PubMed] [Google Scholar]
- 5.Siegal D, Yudin J, Kaatz S, Douketis JD, Lim W and Spyropoulos AC. Periprocedural heparin bridging in patients receiving vitamin K antagonists: systematic review and meta-analysis of bleeding and thromboembolic rates. Circulation 2012;126:1630–1639. [DOI] [PubMed] [Google Scholar]
- 6.Douketis JD, Spyropoulos AC, Kaatz S, Becker RC, Caprini JA, Dunn AS, Garcia DA, Jacobson A, Jaffer AK, Kong DF, Schulman S, Turpie AG, Hasselblad V and Ortel TL. Perioperative Bridging Anticoagulation in Patients with Atrial Fibrillation. N Engl J Med 2015;373:823–833. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Clark NP, Witt DM, Davies LE, Saito EM, McCool KH, Douketis JD, Metz KR and Delate T. Bleeding, Recurrent Venous Thromboembolism, and Mortality Risks During Warfarin Interruption for Invasive Procedures. JAMA Intern Med 2015;175:1163–1168. [DOI] [PubMed] [Google Scholar]
- 8.Moher D, Liberati A, Tetzlaff J and Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. BMJ 2009;339:b2535. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Dunn AS and Turpie AG. Perioperative management of patients receiving oral anticoagulants: a systematic review. Arch Intern Med 2003;163:901–908. [DOI] [PubMed] [Google Scholar]
- 10.Spyropoulos AC, Douketis JD, Gerotziafas G, Kaatz S, Ortel TL and Schulman S. Periprocedural antithrombotic and bridging therapy: recommendations for standardized reporting in patients with arterial indications for chronic oral anticoagulant therapy. J Thromb Haemost 2012;10:692–694. [DOI] [PubMed] [Google Scholar]
- 11.Selby K, Baumgartner C, Levin TR, Doubeni CA, Zauber AG, Schottinger J, Jensen CD, Lee JK and Corley DA. Interventions to Improve Follow-up of Positive Results on Fecal Blood Tests: A Systematic Review. Ann Intern Med 2017;167:565–575. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Klamroth R, Gottstein S, Essers E and Landgraf H. Bridging with enoxaparin using a half-therapeutic dose regimen: safety and efficacy. Vasa 2010;39:243–248. [DOI] [PubMed] [Google Scholar]
- 13.McBane RD, Wysokinski WE, Daniels PR, Litin SC, Slusser J, Hodge DO, Dowling NF and Heit JA. Periprocedural anticoagulation management of patients with venous thromboembolism. Arterioscler Thromb Vasc Biol 2010;30:442–448. [DOI] [PubMed] [Google Scholar]
- 14.Pengo V, Cucchini U, Denas G, Erba N, Guazzaloca G, La Rosa L, De Micheli V, Testa S, Frontoni R, Prisco D, Nante G, Iliceto S, Italian Federation of Centers for the Diagnosis of T and Management of Antithrombotic T. Standardized low-molecular-weight heparin bridging regimen in outpatients on oral anticoagulants undergoing invasive procedure or surgery: an inception cohort management study. Circulation 2009;119:2920–2927. [DOI] [PubMed] [Google Scholar]
- 15.Guyatt GH, Norris SL, Schulman S, Hirsh J, Eckman MH, Akl EA, Crowther M, Vandvik PO, Eikelboom JW, McDonagh MS, Lewis SZ, Gutterman DD, Cook DJ and Schunemann HJ. Methodology for the development of antithrombotic therapy and prevention of thrombosis guidelines: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012;141:53S–70S. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Guyatt GH, Oxman AD, Vist G, Kunz R, Brozek J, Alonso-Coello P, Montori V, Akl EA, Djulbegovic B, Falck-Ytter Y, Norris SL, Williams JW Jr., Atkins D, Meerpohl J and Schunemann HJ. GRADE guidelines: 4. Rating the quality of evidence--study limitations (risk of bias). J Clin Epidemiol 2011;64:407–415. [DOI] [PubMed] [Google Scholar]
- 17.Nyaga VN, Arbyn M and Aerts M. Metaprop: a Stata command to perform meta-analysis of binomial data. Arch Public Health 2014;72:39. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Sjogren V, Grzymala-Lubanski B, Renlund H, Svensson PJ and Sjalander A. Safety and Efficacy of Bridging With Low-Molecular-Weight Heparin During Temporary Interruptions of Warfarin: A Register-Based Cohort Study. Clin Appl Thromb Hemost 2017;23:961–966. [DOI] [PubMed] [Google Scholar]
- 19.Guyatt G, Oxman AD, Akl EA, Kunz R, Vist G, Brozek J, Norris S, Falck-Ytter Y, Glasziou P, DeBeer H, Jaeschke R, Rind D, Meerpohl J, Dahm P and Schunemann HJ. GRADE guidelines: 1. Introduction-GRADE evidence profiles and summary of findings tables. J Clin Epidemiol 2011;64:383–394. [DOI] [PubMed] [Google Scholar]
- 20.Douketis JD, Woods K, Foster GA and Crowther MA. Bridging anticoagulation with low-molecular-weight heparin after interruption of warfarin therapy is associated with a residual anticoagulant effect prior to surgery. Thromb Haemost 2005;94:528–531. [DOI] [PubMed] [Google Scholar]
- 21.Ahmed I, Gertner E, Nelson WB, House CM, Dahiya R, Anderson CP, Benditt DG and Zhu DW. Continuing warfarin therapy is superior to interrupting warfarin with or without bridging anticoagulation therapy in patients undergoing pacemaker and defibrillator implantation. Heart Rhythm 2010;7:745–749. [DOI] [PubMed] [Google Scholar]
- 22.Bajkin BV, Popovic SL and Selakovic SD. Randomized, prospective trial comparing bridging therapy using low-molecular-weight heparin with maintenance of oral anticoagulation during extraction of teeth. J Oral Maxillofac Surg 2009;67:990–995. [DOI] [PubMed] [Google Scholar]
- 23.Birnie DH, Healey JS and Essebag V. Device surgery without interruption of anticoagulation. N Engl J Med 2013;369:1571–1572. [DOI] [PubMed] [Google Scholar]
- 24.Breen DT, Chavalertsakul N, Paul E, Gruen RL and Serpell J. Perioperative complications in patients on low-molecular-weight heparin bridging therapy. ANZ J Surg 2016;86:167–172. [DOI] [PubMed] [Google Scholar]
- 25.Burbury KL, Milner A, Snooks B, Jupe D and Westerman DA. Short-term warfarin reversal for elective surgery--using low-dose intravenous vitamin K: safe, reliable and convenient*. Br J Haematol 2011;154:626–634. [DOI] [PubMed] [Google Scholar]
- 26.Dunn AS, Spyropoulos AC and Turpie AG. Bridging therapy in patients on long-term oral anticoagulants who require surgery: the Prospective Peri-operative Enoxaparin Cohort Trial (PROSPECT). J Thromb Haemost 2007;5:2211–2218. [DOI] [PubMed] [Google Scholar]
- 27.Eisele R, Melzer N, Englert C, Bramlage P and Spannagl M. Bridging with the Low molecular weight heparin certoparin in patients requiring temporary discontinuation of oral anticoagulation - the non-interventional, retrospective REMEMBER study. Thromb Res 2012;130:788–792. [DOI] [PubMed] [Google Scholar]
- 28.Fernandez ON and Mavri A. Evaluation of periprocedural anticoagulation management in patients on long term oral anticoagulants. J Thromb Haemost 2009;7:1071–1072. [Google Scholar]
- 29.Garcia DA, Regan S, Henault LE, Upadhyay A, Baker J, Othman M and Hylek EM. Risk of thromboembolism with short-term interruption of warfarin therapy. Arch Intern Med 2008;168:63–69. [DOI] [PubMed] [Google Scholar]
- 30.Hammerstingl C, Omran H, Tripp C and Poetzsch B. How useful is determination of anti-factor Xa activity to guide bridging therapy with enoxaparin? A pilot study. Thromb Haemost 2009;101:325–332. [PubMed] [Google Scholar]
- 31.Hwang HG, Koo SM, Uh ST and Kim YK. The Perioperative Management of Antithrombotic Therapies Using Enoxaparin. J Korean Med Sci 2017;32:942–947. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Jaffer AK, Ahmed M, Brotman DJ, Bragg L, Seshadri N, Qadeer MA and Klein A. Low-molecular-weight-heparins as periprocedural anticoagulation for patients on long-term warfarin therapy: a standardized bridging therapy protocol. J Thromb Thrombolysis 2005;20:11–16. [DOI] [PubMed] [Google Scholar]
- 33.Majluf Cruz A, Moreno-Hernandez M, Garcia-Chavez J, Vela-Ojeda J, Hernandez-Juarez J and Coria-Ramirez E. Bridging with rivaroxaban in patients with chronic use of vitamin K antagonists. Eur Heart J 2011;32:414. [Google Scholar]
- 34.Malato A, Saccullo G, Lo Coco L, Caramazza D, Abbene I, Pizzo G, Casuccio A and Siragusa S. Patients requiring interruption of long-term oral anticoagulant therapy: the use of fixed sub-therapeutic doses of low-molecular-weight heparin. J Thromb Haemost 2010;8:107–113. [DOI] [PubMed] [Google Scholar]
- 35.Saccullo G, Malato A, Raso S, Santoro M, Zammit V, Casuccio A and Siragusa S. Cancer patients requiring interruption of long-term warfarin because of surgery or chemotherapy induced thrombocytopenia: the use of fixed sub-therapeutic doses of low-molecular weight heparin. Am J Hematol 2012;87:388–391. [DOI] [PubMed] [Google Scholar]
- 36.Santamaria A, Ugarriza A, Munoz C, De Diego I, Lopez-Chulia F, Benet C, Martinez-Gonzalez J, Gomez N, Pina E, Ortin X, Marco P, Roncales FJ and Fontcuberta J. Bemiparin versus unfractionated heparin as bridging therapy in the perioperative management of patients on vitamin K antagonists: the BERTA study. Clin Drug Investig 2013;33:921–928. [DOI] [PubMed] [Google Scholar]
- 37.Skeith L, Taylor J, Lazo-Langner A and Kovacs MJ. Conservative perioperative anticoagulation management in patients with chronic venous thromboembolic disease: a cohort study. J Thromb Haemost 2012;10:2298–2304. [DOI] [PubMed] [Google Scholar]
- 38.Spyropoulos AC, Jenkins P and Bornikova L. A disease management protocol for outpatient perioperative bridge therapy with enoxaparin in patients requiring temporary interruption of long-term oral anticoagulation. Pharmacotherapy 2004;24:649–658. [DOI] [PubMed] [Google Scholar]
- 39.Spyropoulos AC, Turpie AG, Dunn AS, Spandorfer J, Douketis J, Jacobson A, Frost FJ and Investigators R. Clinical outcomes with unfractionated heparin or low-molecular-weight heparin as bridging therapy in patients on long-term oral anticoagulants: the REGIMEN registry. J Thromb Haemost 2006;4:1246–1252. [DOI] [PubMed] [Google Scholar]
- 40.Wilson SJ-A, Morgan J, Gray L, Newman V and Anderson DR. A model for perioperative outpatient management of anticoagulation in high-risk patients: an evaluation of effectiveness and safety. The Canadian Journal of Hospital Pharmacy 2001;54. [Google Scholar]
- 41.Wiszniewski A, Szopinski P, Ratajczak J, Bilski R and Bykowska K. Perioperative bridging therapy with low molecular weight heparin for patients with inherited thrombophilia and antiphospholipid syndrome on long-term acenokumarol therapy. Blood Coagul Fibrinolysis 2011;22:34–39. [DOI] [PubMed] [Google Scholar]
- 42.Constans M, Santamaria A, Mateo J, Pujol N, Souto JC and Fontcuberta J. Low-molecular-weight heparin as bridging therapy during interruption of oral anticoagulation in patients undergoing colonoscopy or gastroscopy. Int J Clin Pract 2007;61:212–217. [DOI] [PubMed] [Google Scholar]
- 43.Spyropoulos AC, Al-Badri A, Sherwood MW and Douketis JD. Periprocedural management of patients receiving a vitamin K antagonist or a direct oral anticoagulant requiring an elective procedure or surgery. J Thromb Haemost 2016;14:875–885. [DOI] [PubMed] [Google Scholar]
- 44.Kowalewski M, Suwalski P, Raffa GM, Slomka A, Kowalkowska ME, Szwed KA, Borkowska A, Kowalewski J, Malvindi PG, Undas A, Windyga J, Pawliszak W, Anisimowicz L, Carrel T, Paparella D and Lip GY. Meta-analysis of uninterrupted as compared to interrupted oral anticoagulation with or without bridging in patients undergoing coronary angiography with or without percutaneous coronary intervention. Int J Cardiol 2016;223:186–194. [DOI] [PubMed] [Google Scholar]
- 45.Longstreth WT Jr., Bernick C, Fitzpatrick A, Cushman M, Knepper L, Lima J and Furberg CD. Frequency and predictors of stroke death in 5,888 participants in the Cardiovascular Health Study. Neurology 2001;56:368–375. [DOI] [PubMed] [Google Scholar]
- 46.Durrleman N, Pellerin M, Bouchard D, Hebert Y, Cartier R, Perrault LP, Basmadjian A and Carrier M. Prosthetic valve thrombosis: twenty-year experience at the Montreal Heart Institute. J Thorac Cardiovasc Surg 2004;127:1388–1392. [DOI] [PubMed] [Google Scholar]
- 47.Carrier M, Le Gal G, Wells PS and Rodger MA. Systematic review: case-fatality rates of recurrent venous thromboembolism and major bleeding events among patients treated for venous thromboembolism. Ann Intern Med 2010;152:578–589. [DOI] [PubMed] [Google Scholar]
- 48.Barnes GD, Kurlander J, Haymart B, Kaatz S, Saini S and Froehlich JB. Bridging Anticoagulation Before Colonoscopy: Results of a Multispecialty Clinician Survey. JAMA Cardiol 2016;1:1076–1077. [DOI] [PubMed] [Google Scholar]
- 49.Barnes GD and Mouland E. Peri-Procedural Management of Oral Anticoagulants in the DOAC Era. Prog Cardiovasc Dis 2018. [DOI] [PMC free article] [PubMed]
- 50.Beyer-Westendorf J, Gelbricht V, Forster K, Ebertz F, Kohler C, Werth S, Kuhlisch E, Stange T, Thieme C, Daschkow K and Weiss N. Peri-interventional management of novel oral anticoagulants in daily care: results from the prospective Dresden NOAC registry. Eur Heart J 2014;35:1888–1896. [DOI] [PubMed] [Google Scholar]
- 51.Sherwood MW, Douketis JD, Patel MR, Piccini JP, Hellkamp AS, Lokhnygina Y, Spyropoulos AC, Hankey GJ, Singer DE, Nessel CC, Mahaffey KW, Fox KA, Califf RM and Becker RC. Outcomes of temporary interruption of rivaroxaban compared with warfarin in patients with nonvalvular atrial fibrillation: results from the rivaroxaban once daily, oral, direct factor Xa inhibition compared with vitamin K antagonism for prevention of stroke and embolism trial in atrial fibrillation (ROCKET AF). Circulation 2014;129:1850–1859. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.