Estimating thromboembolic risk

Discontinuing antithrombotic therapy to avoid the risk of major bleeding during surgery is common practice around the world. However, understanding the risks of stopping antithrombotic medication is paramount to surgical practice, as the risk of clinical harm due to thrombotic events is believed to be higher than those of major perioperative bleeding (figure 1). 

Fatality and mortality rates due to thrombotic and major bleeding events

Figure 1. Fatality and mortality rates due to thrombotic and major bleeding events (Deviri et al., 1991; Feigin et al., 2014; Wu et al., 2014; Wu et al., 2015).
VTE, venous thromboembolism

This is why the American College of Surgeons’ (ACS) guidelines recommend assessing thromboembolic risk as the first step in developing a perioperative antithrombotic management strategy (Hornor et al., 2018).

The ACS guidelines have stratified thromboembolic risk into high (>10%), moderate (5–10%), and low (<5%) categories and are adapted from the 9th Edition of the American College of Chest Physicians’ (CHEST) guidelines (Douketis et al., 2012).

Thromboembolic risk should be considered on an individual basis and is dependent on several patient factors, the most important of which is the medical indication of the antithrombotic agent. These guidelines discuss the risks of several conditions that require the use of chronic anticoagulation, including nonvalvular atrial fibrillation, prosthetic heart valves, venous thromboembolism, coronary artery disease, stroke, and peripheral arterial disease.

Nonvalvular atrial fibrillation

The CHA2DS2-VASc score is a tool often used to predict the likelihood of a stroke occurring in patients with non-rheumatic atrial fibrillation. It is recommended that the calculated score be used to categorise patients into low, moderate or high thromboembolic risk (figure 2) (Friberg et al., 2012).

The CHA2DS2-VASc score can be stratified into three categories of thromboembolic risk

Figure 2. The CHA2DS2-VASc score can be stratified into three categories of thromboembolic risk (Hornor et al., 2018).

It is important to note that the CHA2DS2-VASc score hasn’t been prospectively validated in the perioperative setting, however, it has been used to estimate thromboembolic risk if anticoagulation is held and to determine whether heparin bridging is indicated perioperatively.

Prosthetic heart valves

When assessing the perioperative thromboembolic risk in patients with prosthetic heart valves, there are a number of elements to take into consideration, such as the

  • type
  • number
  • location of the valvular prosthetic
  • as well as the presence of additional cardiac risk factors
    • such as atrial fibrillation, previous stroke or transient ischemic attack, hypertension, diabetes, congestive heart failure, age >75 years (Whitlock et al., 2012)


Thromboembolic risk when anticoagulated with warfarin according to prosthetic heart valve type

Figure 3. Thromboembolic risk when anticoagulated with warfarin according to prosthetic heart valve type (Cannegieter et al., 1994; Whitlock et al., 2012; Hornor et al., 2018).

Bioprosthetic valves

Bioprosthetic heart valves do not require long-term anticoagulation for the valve alone; many guidelines recommend short-term anticoagulation after implantation (3–6 months) to decrease the incidence of thromboembolic events. If a patient is receiving anticoagulation after this period, then it is likely for a secondary reason and this should be factored into a risk/benefit analysis for stopping anticoagulation (Vahanian et al., 2012; Whitlock et al., 2012).

Elective operations should be delayed 3–6 months after implantation of a bioprosthetic valve.

In the case of nonelective operations, the surgeon should have no hesitancy stopping warfarin without any bridging methodology and restarting anticoagulation when there is no increased risk of bleeding (Whitlock et al., 2012).

Venous thromboembolism

Premature cessation of anticoagulation in the first three months after an acute venous thromboembolism (VTE) event has been shown to be associated with a 1-month risk of recurrent VTE of 40%, and a 10% risk for the subsequent 2 months. Annual risk of VTE falls to 15% after 3 months of anticoagulation (Research Committee of the British Thoracic Society, 1992; Levine et al., 1995). It is therefore recommended that elective surgery is postponed for 3 months following a diagnosis of VTE (Douketis et al., 2012).

The risk of VTE reoccurrence is also dependent of whether the VTE was provoked or unprovoked, and if there are any underlying risk factors such as cancer or inherited thrombophilia (de Jong et al., 2012). A provoked VTE is classified as an identifiable factor such as an inciting event (surgery, a long flight, hospital admission), transient major risk factors (major surgery, >30 minutes hospitalisation or immobility, pregnancy, oestrogen therapy), or permanent risk factors (inherited thrombophilia, chronic heart failure, inflammatory bowel disease, malignancy) (Kearon et al., 2016).

Thromboembolic risk in patients with a history of VTE

Figure 4. Thromboembolic risk in patients with a history of VTE (Bates et al., 2012; Hornor et al., 2018).
VTE, Venous thromboembolism

The ACS guidelines recommends close consultation with the prescribing physician when developing a perioperative anticoagulation strategy for patients with a history of an unprovoked VTE or a provoked/unprovoked life-threatening pulmonary embolism (de Jong et al., 2012; Hornor et al., 2018).

Coronary artery disease

For some patients with coronary artery disease, stent implantation via surgery is the most effective option. However, following stent implantation patients are at risk of stent thrombosis, with the risk being highest in the following 4–6. One strong risk factor for stent thrombosis is the discontinuation of dual antiplatelet therapy (Iakovou et al., 2005).

It is currently recommended that dual agent antiplatelet therapy is prescribed for at least 6–12 months after placement of bare-metal stents and drug-eluting stents (Kleiman, 2012). Therefore, when considering the discontinuation of antiplatelet therapy prematurely in patients with coronary artery disease, the risk of cardiac complications needs to be carefully weighed against major surgical bleeding risk. This is a consideration often needed as 5% of patients with coronary stents require non-cardiac surgery within 1 year of coronary stent implantations, rising to 23% within 5 years (Savonitto et al., 2011).

Recommended delays in elective surgery following stent placement

Figure 5. Recommended delays in elective surgery following stent placement (Fleisher et al., 2014; Hornor et al., 2018).

If surgery can’t be deferred a discussion should be had with the prescribing cardiologist regarding the risk/benefit ratio of stopping antiplatelet therapy and patient disclosure should be documented (Fleisher et al., 2014).


Recent history of stroke has been significantly associated with major cardiovascular events after elective non-cardiac surgery (stroke within 3 months, OR 14.23; 95% CI 11.61–17.45). However, the risk plateaued at 9 months after a stroke (Jørgensen et al., 2014) and, therefore, elective surgery is recommended to be deferred at least 9 months after ischaemic stroke (Hornor et al., 2018).

Peripheral arterial disease

Pharmacological treatment of peripheral arterial disease (PAD) with dual antiplatelet therapy is recommended for patients who have ischaemic limb symptoms and clinically significant coronary or cerebrovascular disease. For patients with PAD without clinically significant coronary or cardiovascular disease a single agent such as aspirin, clopidogrel, or ticagrelor is recommended (Gerhard-Herman et al., 2017).

 Patients with PAD considered high risk for perioperative thromboembolism

Figure 6. Patients with PAD considered high risk for perioperative thromboembolism (Hornor et al., 2018).

Antithrombotic medication should be managed accordingly in those considered high risk, and patients who receive antithrombotic medications for symptomatic PAD should be managed in close consultation with a vascular specialist or vascular surgeon.

The underlying health issues associated with antithrombotic complications play a major factor in assessing risk, but procedural risk also has a large influence on the chosen management strategies. Learn more about procedural risks in the next section of the ACS guidelines.