To Bridge or Not to Bridge: Modelling Periprocedural Anticoagulation Management

Purpose: For atrial fibrillation (AF) patients receiving vitamin K antagonists (VKAs), careful management of anticoagulation is important around surgical procedures to minimize the stroke and bleeding risks. If the VKA needs to be stopped periprocedurally to reduce the risk of bleeding, a decision needs to be made whether to bridge this period with a low-molecular weight heparin (LMWH). We aimed to develop a model to compare two periprocedural strategies for AF patients that have to interrupt VKA treatment: administering a LMWH or forgoing bridging therapy. Method(s): A probabilistic Markov model was developed to simulate both a bridge and a non-bridge cohort of AF patients periprocedurally. Modelled events were based on the clinically used CHA2DS2-VASc and HAS-BLED stroke and bleeding prediction rules. To predict strokes, INR values were considered. Quality-adjusted life expectancy, based on the beforementioned clinical endpoints, was the main outcome considered. Result(s): The base case analysis shows that bridging anticoagulation increases the bleeding rate, but reduces the stroke rate. Bridging may be beneficial for patients with a CHA2DS2-VASc scores of 6 or higher and HAS-BLED scores of 0 to 2. For expected shorter periods to reach therapeutic INR, bridging therapy is less likely to be beneficial. Conclusion(s): For patients at high risk of bleeding, bridging anticoagulation Is not likely to be beneficial. For patients at high risk of stroke and low risk of bleeding, bridging anticoagulation may result in additional quality adjusted life years. INR management is an important factor to consider periprocedurally when making the decision whether to bridge.