Over 40 million Americans regularly use aspirin and 5 million people are on long-term anticoagulant therapy. The common use of antithrombotic medications, including both antiplatelets and anticoagulants, has distinct perioperative implications. Historically, aspirin, warfarin, and clopidogrel were the only antithrombotics available for outpatient therapy, but in recent years a plethora of new agents have been introduced. These include direct thrombin inhibitors (dabigatran), direct factor Xa inhibitors (rivaroxaban, apixaban), and ADP receptor antagonists (ticlopidine, prasugrel, ticagrelor), among others ( Table ). Use of these drugs is becoming more prevalent because of their efficacy and because they do not require serial monitoring of coagulation parameters.

Vitreoretinal surgeons are often faced with the dilemma of deciding whether to continue or suspend antithrombotic therapy prior to surgery. Although preoperative discontinuation may diminish the potential for perioperative ocular hemorrhage, it may also increase the probability for systemic complications such as stroke and myocardial infarction. In this regard, it is useful to assign individual patient and procedural risk factors according to a high or low level of risk for thromboembolic complications. Assessment of patient risk factors should be based on an internist’s recommendations and consideration should be given to the underlying condition for which the patient is receiving anticoagulants. For example, patients prophylactically taking antithrombotics for atherosclerosis or a family history of heart disease may be candidates for interval suspension of therapy. Conversely, patients receiving treatment for conditions including the recent placement of a drug-eluting stent, new-onset atrial fibrillation, mechanical heart valves, deep vein thrombosis, or pulmonary embolism may be at substantially greater risk of thrombosis if therapy is interrupted.

Hemorrhagic risks associated with vitreoretinal surgery include hyphema and subretinal, suprachoroidal, and vitreous hemorrhages. Prevalence of hemorrhagic complications may vary among different systemic conditions and their effect on vitreoretinal anatomy. For instance, a vascular disease such as diabetic retinopathy poses significantly different risks as compared to rhegmatogenous retinal detachment repair. One may logically infer that these complications are more prevalent in patients taking antithrombotic agents; however, this assumption is not consistently supported by the literature. Some studies failed to demonstrate higher rates of intraoperative or postoperative bleeding in patients receiving anticoagulants, while another reported the converse. Similarly, platelet inhibition has not been conclusively proven to increase the risk of vitreoretinal surgical bleeding events. The clinical significance of hemorrhage associated with platelet inhibition has, for the greater part, been minimal, with spontaneous resolution and no long-term sequelae. Newer surgical techniques, such as microincisional surgery, and shorter operating times may further reduce the incidence of perioperative bleeding and should diminish the importance attached to discontinuation of antithrombotic therapy.

Posterior segment surgery is most commonly performed under regional anesthesia with a peribulbar or retrobulbar block. Hemorrhagic complications associated with these blocks are rare and range from mild periorbital ecchymosis and subconjunctival hemorrhage to sight-threatening retrobulbar bleeds. The incidence of retrobulbar hemorrhage following an ophthalmic block is exceptionally low and rarely of any long-term adverse functional significance. Although antithrombotics would theoretically be expected to amplify bleeding from a block, there are no evidence-based data to substantiate this; one study found that the risk was the same whether these drugs were held or sustained.

Current general medical practice favors the perioperative continuation of anticoagulation with most surgical procedures. This stems from a heightened awareness of the risks and complications associated with interruption of these medications, and an appreciation of what constitutes a procedure that is at high risk for hemorrhage. In a recent article from the Annals of Surgery , Gerstein and associates advocate for the continuation of aspirin during the perioperative period, but make an understandable exception for procedures where “any additional excessive blood loss would lead to worse outcomes related to morbidity or mortality.” Of note, the authors consider posterior segment eye surgery to be one such procedure, ranking among intracranial and spinal canal surgeries. In another recent review in the New England Journal of Medicine , Baron and associates also categorize vitreoretinal surgery as “high-risk” for bleeding complications. They base this on statements published by the International Society on Thrombosis and Haemostasis, who classify any hemorrhage as “major” if it is “fatal, cause[s] hemodynamic instability, or require[s] a second [surgical] intervention.” This definition also encompasses any bleeding that “occurs in a critical area or organ,” inclusive of the eye.

Hemorrhage may occur during any intraocular surgery. Cataract surgery is considered “low-risk” because it typically involves avascular ocular structures; vitreoretinal surgery commonly involves the manipulation of vascular tissue that is intrinsically susceptible to bleeding. The question of whether vitreoretinal surgery should be considered a high- or low-risk procedure is an empirical one. Although Gerstein and Baron, both non-ophthalmologists, categorize posterior segment eye surgery as high-risk, the preponderance of current ophthalmology literature suggests it is not. Moreover, it has been shown that most hemorrhagic complications arising from intraocular surgery typically do not cause chronic sequelae or consume major health-care resources. This implies that, for vitreoretinal surgery, the preoperative continuation of antithrombotic treatment may be advantageous because hemorrhagic complications tend to have minimal effect on visual outcome while systemic anticoagulation substantially impacts morbidity and long-term mortality. Nonetheless, patients using these medications at the time of surgery do have a low but definite risk of bleeding. Physicians should weigh the risks and benefits of these agents and use their best clinical judgment for each individual patient. For these reasons, the classification of posterior segment surgery as high-risk needs further evidence-based studies, and perhaps a change to moderate- or low-risk is warranted.