Perioperative Anticoagulation: 3 Key Approaches Compared
When warfarin or a DOAC must pause for surgery, the evidence now favors simple interruption and resumption over routine heparin bridging.
One-Minute Takeaway
- Perioperative anticoagulation in most patients no longer needs heparin bridging; randomized trial evidence — the highest tier available here — supports interrupting and resuming warfarin or a DOAC on a simple, fixed schedule.
- In the BRIDGE randomized trial, forgoing bridging markedly reduced major bleeding versus low-molecular-weight heparin (RR 0.41, 95% CI 0.20–0.78), with no increase in arterial thromboembolism.
- The PAUSE cohort managed DOACs with no bridging and no routine coagulation testing; 30-day major bleeding stayed under 2% and stroke under 1% across apixaban, dabigatran, and rivaroxaban.
- For pacemaker or defibrillator surgery, continued warfarin cut clinically significant pocket hematoma to 3.5% versus 16.0% with bridging (RR 0.19, 95% CI 0.10–0.36).
- The 2022 CHEST guideline gives a strong recommendation against bridging in atrial fibrillation and for continuing warfarin during device implantation.
Why Perioperative Anticoagulation Matters
Roughly one in six patients with atrial fibrillation who take an oral anticoagulant will need an elective surgery or procedure each year. On a global scale that is several million people annually facing a temporary anticoagulation decision.
For two decades, many of these patients received heparin “bridging” while their warfarin was held. The logic was intuitive: keep the patient anticoagulated for as much of the perioperative window as possible to prevent stroke.
That intuition has not survived contact with randomized data. The strategy chosen around a procedure changes bleeding risk substantially while doing little to change the already-low risk of arterial thromboembolism. Getting the choice right is therefore a safety issue, not a formality.
The Clinical Problem in Perioperative Anticoagulation
Every perioperative decision balances two competing harms. Continuing full anticoagulation through an invasive procedure raises bleeding risk. Stopping it opens a window in which a susceptible patient could, in theory, have a stroke or systemic embolism.
Bridging was designed to have it both ways: interrupt the long-acting drug, but substitute a short-acting heparin so the anticoagulation-free window is minimized. The unexamined assumption was that this narrow window is where perioperative strokes come from.
Two variables drive the decision. The first is thrombotic risk, estimated from scores such as CHADS₂ or CHA₂DS₂-VASc and from features like a mechanical valve or a recent thromboembolic event. The second is the bleeding risk of the specific procedure.
The pivotal insight from randomized evidence is that the perioperative stroke rate in most anticoagulated patients is low regardless of bridging — while bridging reliably adds bleeding. The rest of this review examines how each strategy performs against those two outcomes.
Available Approaches
Three strategies dominate practice: interrupt the anticoagulant without any heparin substitute, interrupt it but bridge with heparin, or continue the anticoagulant through a low-risk procedure. The table below summarizes the trials and pooled analyses that define each, drawn only from sources retrieved and confirmed for this review.
| Study / Source | Design | N | Population | Comparison | Key result |
|---|---|---|---|---|---|
| Siegal 2012V | Systematic review & meta-analysis (34 studies) | 12,278 | VKA-treated, mainly AF and heart valves | Heparin bridging vs no bridging | Thromboembolism no different (OR 0.80, 95% CI 0.42–1.54); major bleeding markedly higher with bridging (OR 3.60) |
| BRIDGE 2015V | RCT, double-blind, placebo-controlled | 1,884 | AF on warfarin, CHADS₂ ≥1, no mechanical valve | No bridging vs LMWH (dalteparin) bridging | Major bleeding 1.3% vs 3.2% (RR 0.41, 95% CI 0.20–0.78); arterial thromboembolism 0.4% vs 0.3% (noninferior) |
| PERIOP2 2021V | RCT, double-blind, placebo-controlled | 1,471 | AF or mechanical heart valve, warfarin interrupted | Postoperative dalteparin bridging vs placebo | Major thromboembolism 1.0% vs 1.2% (risk difference −0.3%, 95% CI −1.3 to 0.8); no benefit from bridging |
| BRUISE CONTROL 2013V | RCT, single-blind | 681 | High thrombotic risk (≥5%/yr), device surgery | Continued warfarin vs heparin bridging | Clinically significant pocket hematoma 3.5% vs 16.0% (RR 0.19, 95% CI 0.10–0.36) |
| Essebag 2019 (BRUISE CONTROL 1+2 pooled)V | Pooled analysis of two RCTs | 1,343 | Device surgery on warfarin or a DOAC | Continued vs interrupted DOAC; antiplatelet effect | DOAC continue vs interrupt: hematoma 2.1% vs 2.1% (no difference); concomitant antiplatelet doubled hematoma risk |
| PAUSE 2019V | Prospective cohort management study | 3,007 | AF on apixaban, dabigatran, or rivaroxaban | Standardized interruption, no bridging, no routine testing | 30-day major bleeding 0.90–1.85%; arterial thromboembolism 0.16–0.60% by drug |
Interruption Without Bridging
This is the strategy the evidence supports most directly. For warfarin, treatment is typically stopped about five days before the procedure so the anticoagulant effect can wane, then resumed within roughly 24 hours once hemostasis is secured.
The BRIDGE trial randomized 1,884 patients with atrial fibrillation and no mechanical valve to bridging or matching placebo. Forgoing bridging led to less major bleeding (1.3% vs 3.2%) and remained noninferior for arterial thromboembolism, which was low in both arms.
For direct oral anticoagulants, the short half-life changes the mechanics. PAUSE managed 3,007 patients on apixaban, dabigatran, or rivaroxaban by omitting the drug for one day before a low-bleeding-risk procedure and two days before a high-bleeding-risk one, with longer intervals for dabigatran and reduced kidney function.
In PAUSE, this cohort was managed without heparin bridging and without routine preoperative coagulation testing. Rates of major bleeding stayed under 2% and arterial thromboembolism under 1% in each drug group, and more than 90% of patients had minimal or no residual anticoagulant at surgery.
Interruption With Heparin Bridging
Bridging substitutes a short-acting heparin during the warfarin-free window. The Siegal meta-analysis of 34 studies and 12,278 patients found that bridged patients had a similar thromboembolism rate to non-bridged patients (OR 0.80, 95% CI 0.42–1.54) but substantially more bleeding.
PERIOP2 tested the question in higher-risk patients that BRIDGE excluded, enrolling people with atrial fibrillation or a mechanical heart valve. Postoperative dalteparin bridging did not reduce major thromboembolism compared with placebo (1.0% vs 1.2%), and offered no signal of benefit in either subgroup.
Together these findings reframe bridging as a strategy that adds bleeding risk without a matching reduction in clot risk for most patients. Its role, if any, narrows to a small group at the highest thrombotic risk, where the trade-off must be individualized.
Continued Anticoagulation Through the Procedure
For some low-bleeding-risk operations, not stopping at all is an option. BRUISE CONTROL randomized 681 patients at high thrombotic risk undergoing pacemaker or defibrillator surgery to continued warfarin or heparin bridging.
Continued warfarin reduced clinically significant device-pocket hematoma to 3.5%, compared with 16.0% with bridging (RR 0.19, 95% CI 0.10–0.36). The trial was stopped early for benefit. Investigators proposed an “anticoagulant stress test” effect: bleeding that surfaces intraoperatively while the wound is open can be controlled before it becomes a hematoma.
The pooled BRUISE CONTROL 1 and 2 analysis extended the picture to direct oral anticoagulants, where continuing versus interrupting the DOAC produced the same low hematoma rate (2.1% in both). That analysis also found concomitant antiplatelet therapy doubled the hematoma risk, a reminder that co-medication often matters more than the anticoagulant itself.
Clinical pearl. For a pacemaker or defibrillator implant in an anticoagulated patient, the randomized evidence favors continuing warfarin over bridging. For a DOAC, both continuing and briefly interrupting produced equally low hematoma rates — so the decision can hinge on procedure logistics and concomitant antiplatelet use rather than on bridging.
Comparative Analysis
Read together, these studies point in one direction across drug classes and risk strata. Whether the anticoagulant is warfarin or a DOAC, and whether the patient has atrial fibrillation or a mechanical valve, adding heparin bridging consistently increases bleeding without a corresponding reduction in thromboembolism.
The bleeding signal is not subtle. Bridging nearly tripled major bleeding in BRIDGE and produced a fivefold-higher hematoma rate in device surgery in BRUISE CONTROL. The Siegal meta-analysis put the pooled major-bleeding odds ratio for bridging at 3.60.
The thromboembolic side of the ledger is quiet. Arterial thromboembolism rates clustered at or below roughly 1% whether or not patients were bridged, in the randomized trials and in the large DOAC cohort alike. Bridging did not move that number.
Resumption timing is the other half of the strategy. Across BRIDGE and PAUSE, anticoagulation was restarted within about a day after low-bleeding-risk procedures and after two to three days for high-bleeding-risk ones, once hemostasis was secure. The recurring theme is that a short, standardized interruption — not a heparin scaffold — is what keeps both risks low.
Quality and Consistency of the Evidence
The strength of these conclusions varies by outcome and by population. The list below rates the evidence supporting each claim.
- Bridging increases bleeding — High
Consistent across a double-blind RCT (BRIDGE), a device-surgery RCT (BRUISE CONTROL), and a large meta-analysis, all pointing the same way. - Skipping bridging does not raise stroke risk — Moderate
Supported by randomized noninferiority data, but perioperative event rates were low, which limits precision for the highest-risk patients. - Standardized DOAC interruption is safe — Moderate
PAUSE is large and rigorous but is a single-arm cohort, not a randomized comparison, so it establishes safety rather than superiority. - Applicability to mechanical valves and very high-risk patients — Low
BRIDGE excluded mechanical valves; PERIOP2 included them but with modest event numbers, leaving these subgroups less certain.
What the Evidence Does Not Show
Specific, named gaps remain.
Patients undergoing high-bleeding-risk major surgery — cardiac, intracranial, spinal, or major cancer operations — were largely excluded from BRIDGE, so its bleeding and thromboembolism estimates should not be extended to them.
The very highest thrombotic-risk groups are thinly represented. Few patients with a CHADS₂ score of 5 or 6, older-generation mechanical mitral valves, or a stroke within the prior three months were studied, so the residual case for selective bridging in those patients rests on weaker data.
PAUSE did not include edoxaban, and its findings do not transfer automatically to that agent. It also enrolled a predominantly white population, and did not address patients anticoagulated for venous thromboembolism rather than atrial fibrillation.
Optimal resumption timing after high-bleeding-risk surgery, and the management of neuraxial anesthesia in DOAC users, remain areas where the evidence guides but does not fully settle practice.
Practical Implications
The evidence supports a default of simple interruption and resumption for most patients, with bridging reserved as an individualized exception. The points below describe what the data support, not fixed directives; local protocols and patient-specific factors still govern the final decision.
For atrial fibrillation on warfarin, the evidence supports interrupting without bridging, stopping about five days before the procedure and resuming within roughly a day once hemostasis is secure. The 2022 CHEST guideline makes this a strong recommendation.
For a DOAC, the supported approach is a short, half-life-based interruption — about one day before a low-bleeding-risk procedure and two before a high-risk one, extended for dabigatran with reduced kidney function — without bridging or routine coagulation testing.
For pacemaker or defibrillator implantation, continuing warfarin is favored over interruption plus bridging, and is also a strong CHEST recommendation. For device surgery on a DOAC, continuing or briefly interrupting are both reasonable.
Selective bridging may still be considered for a small group at the highest thrombotic risk, such as certain mechanical valves or a very recent thromboembolic event, through shared decision-making. Any drug dosing should be verified against current prescribing information, and concomitant antiplatelet therapy weighed separately, since it independently raises bleeding risk.
Evidence Grade and Bottom Line
Overall grade: Moderate to High- The bleeding benefit of avoiding heparin bridging is supported by high-quality, consistent randomized and pooled evidence.
- The thromboembolic safety of skipping bridging or continuing anticoagulation is graded moderate, constrained by low event rates rather than by contradictory findings.
- Bottom line: for most patients needing a procedure, interrupt and resume anticoagulation on a simple schedule, or continue it for selected low-risk device surgery — bridging adds bleeding without preventing clots.
- Reserve consideration of bridging for the narrow highest-risk group, and treat mechanical-valve and major-surgery scenarios as areas where evidence is thinner and individualization matters most.
Article Information and References
- Douketis JD, Spyropoulos AC, Murad MH, et al. Perioperative Management of Antithrombotic Therapy: An American College of Chest Physicians Clinical Practice Guideline. Chest. 2022;162(5):e207–e243. DOI: https://doi.org/10.1016/j.chest.2022.07.025
- Siegal D, Yudin J, Kaatz S, Douketis JD, Lim W, Spyropoulos AC. Periprocedural heparin bridging in patients receiving vitamin K antagonists: systematic review and meta-analysis of bleeding and thromboembolic rates. Circulation. 2012;126(13):1630–1639. DOI: https://doi.org/10.1161/CIRCULATIONAHA.112.105221
- Douketis JD, Spyropoulos AC, Kaatz S, et al; BRIDGE Investigators. Perioperative bridging anticoagulation in patients with atrial fibrillation. N Engl J Med. 2015;373(9):823–833. DOI: https://doi.org/10.1056/NEJMoa1501035
- Kovacs MJ, Wells PS, Anderson DR, et al; PERIOP2 Investigators. Postoperative low molecular weight heparin bridging treatment for patients at high risk of arterial thromboembolism (PERIOP2): double blind randomised controlled trial. BMJ. 2021;373:n1205. DOI: https://doi.org/10.1136/bmj.n1205
- Birnie DH, Healey JS, Wells GA, et al. Pacemaker or defibrillator surgery without interruption of anticoagulation. N Engl J Med. 2013;368(22):2084–2093. DOI: https://doi.org/10.1056/NEJMoa1302946
- Essebag V, Healey JS, Joza J, et al. Effect of Direct Oral Anticoagulants, Warfarin, and Antiplatelet Agents on Risk of Device Pocket Hematoma: Combined Analysis of BRUISE CONTROL 1 and 2. Circ Arrhythm Electrophysiol. 2019;12(10):e007545. DOI: https://doi.org/10.1161/CIRCEP.119.007545
- Douketis JD, Spyropoulos AC, Duncan J, et al. Perioperative Management of Patients With Atrial Fibrillation Receiving a Direct Oral Anticoagulant. JAMA Intern Med. 2019;179(11):1469–1478. DOI: https://doi.org/10.1001/jamainternmed.2019.2431
- Schulman S, Carrier M, Lee AY, et al; Periop Dabigatran Study Group. Perioperative management of dabigatran: a prospective cohort study. Circulation. 2015;132(3):167–173. DOI: https://doi.org/10.1161/CIRCULATIONAHA.115.015688
- Healey JS, Eikelboom J, Douketis J, et al; RE-LY Investigators. Periprocedural bleeding and thromboembolic events with dabigatran compared with warfarin: results from the RE-LY randomized trial. Circulation. 2012;126(3):343–348. DOI: https://doi.org/10.1161/CIRCULATIONAHA.111.090464
- Sherwood MW, Douketis JD, Patel MR, et al; ROCKET AF Investigators. Outcomes of temporary interruption of rivaroxaban compared with warfarin in patients with nonvalvular atrial fibrillation (ROCKET AF). Circulation. 2014;129(18):1850–1859. DOI: https://doi.org/10.1161/CIRCULATIONAHA.113.005754
- Douketis JD, Spyropoulos AC, Spencer FA, et al. 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(2 Suppl):e326S–e350S. DOI: https://doi.org/10.1378/chest.11-2298