This article is the second in a three-part series highlighting the complex clinical issues that are facing early career cardiologists. John Ryan, MD, FACC, assistant professor at the University of Utah, interviewed Eiman Jahangir, MD, FACC, consultative cardiologist at Ochsner Medical Center, Hansie Mathelier, MD, FACC, assistant professor at the University of Pennsylvania and Ben Freed MD, FACC, assistant professor at Northwestern University.

In the second of three interviews, Dr. Ryan asks the panel, "In patients with atrial fibrillation (AFib) how do you decide whether to bridge patients with anticoagulation and what strategies do you use (low molecular weight heparin [LMWH] or unfractionated heparin [UFH], etc.)?"

Eiman Jahangir: In individuals with AFib, my decision to bridge or not is based on the risk of thromboembolic disease, the risk of bleeding with surgery and individual preference. In individuals with a moderate or high risk of thromboembolic disease, such as a CHA2DS2-VASc of >2 or if they have had prior thromboembolic disease, I will be more likely to bridge, though there is no convincing data from prospective studies supporting this. I will discuss with these patients the risks, benefits and inconveniences of bridging, including the low rates of peri-operative thromboembolism (0.3 percent to 1 percent in the Dresden registry and RE-LY, ROCKET AF, and ARISTOTLE trials). The patient and I will then gauge if it is worth the effort for such a low risk of thromboembolism. Regardless, for all patients I will try to minimize the time anticoagulation is stopped, particularly when surgery would present a low risk for bleeding. In this instance, I will try to continue the anticoagulant as long as the surgeon will allow. The main instance I deviate from this practice is among patients undergoing high-risk surgery. I would discontinue the anticoagulant and recommend against bridging in these patients.

Hansie Mathelier: Regarding bridging strategy for patients with AFib, I don't bridge everyone. My approach is to consider their thromboembolic risks, age, and renal function, as well as the type of surgery.

For patients that are low risk CHADS2 score 0 – 2 and no prior TIA, I wouldn't bridge since their annual risk is less than 5 percent. For patients that are high risk, with annual percent of greater than 10 percent, I would bridge. These are individuals with a CHADS2 score of 5 or 6 or patients with a recent stroke or TIA within the past 3 months. For those patients that are in the moderate risk CHADS2 of 3 or 4, this is more of a gray zone. I also consider the type of surgery and the phenotype of the patient. If the patient was hospitalized pre-surgery, then I would initiate in this subset. If this scenario is for an outpatient, I would use LMWH if there is normal renal function. For patients currently hospitalized, I would use heparin and use a nurse run protocol.

The results of the BRIDGE study also provide a better strategy for these patients. For all patients post surgery, I defer to the surgeon regarding when they feel it's safe to restart anticoagulation. Often I would not bridge and use Coumadin only, particularly in orthopedic surgeries.

Ben Freed: We know from multiple studies that patients with AFib are at a higher risk of thromboembolic disease compared to the general population. We also know that the more cardiovascular risk factors patients have (ie age, congestive heart failure, hypertension, diabetes, stroke and vascular disease), the higher their risk of a thrombotic event. Physicians use this knowledge everyday to help them decide whether or not their patient with AFib should be taking an anticoagulant, aspirin or nothing.

The decision to bridge those patients on oral anti-coagulation with either LMWH or UFH is less clear. Most physicians use either the CHADS2 or CHA2DS2-VASc score for calculating the annualized thromboembolic risk of their patients with AFib. Similarly, physicians will use these risk scores for deciding whether or not to bridge with a low threshold for initiating therapy.

I don't entirely agree with this strategy. The risk scores mentioned above provide annualized risks for a thrombotic event and were not specifically designed to answer the question of whether or not to bridge. Even if you are treating a patient with all the risk factors, the yearly risk is anywhere from 15.2 percent to 18.2 percent (depending on the risk score). That translates to a daily risk of anywhere from 0.04 percent to 0.05 percent which is quite low and might not justify the bleeding risks associated with bridging.

Given this, I do not routinely bridge patients with AFib on anticoagulation. For very high-risk patients (e.g. recent stroke) who will require long-term bridging, I do consider using either LMWH or UFH but, frankly, I am not sure if I am treating the patient or myself.