Perioperative management of antiplatelet therapy in ophthalmic surgery

Abstract

Many patients undergoing ophthalmic surgery are elderly with comorbidities requiring antiplatelet therapy to prevent thromboembolic or atherothrombotic events. The use of antiplatelet therapy has expanded over the years, predisposing these patients to hemorrhagic complications perioperatively. The risk of hemorrhagic complications must be weighed against the risk of thromboembolic events with cessation of antiplatelet therapy. The decision to continue or interrupt antiplatelet therapy in the setting of ophthalmic surgery is based upon various factors, including the type of surgery and each patient’s comorbidities. This review examines the risks of thrombotic complications versus hemorrhagic complications in different types of ophthalmic surgeries with the use of antiplatelet medications and provides evidence-based recommendations regarding perioperative management of antiplatelet therapy

Introduction

Many patients undergoing ocular surgery are elderly with significant comorbidity requiring antithrombotic therapy to reduce their risk of thromboembolic and atherothrombotic events. These drugs are often discontinued preoperatively due to the risk of hemorrhagic complications. However, this risk has to be weighed against the risk to the patient of discontinuing the treatment. Antithrombotics can be subdivided into antiplatelet and anticoagulant drugs. This review will focus on the perioperative management of antiplatelet therapy in ophthalmic surgery.

Types of Antiplatelet Therapy

An increasing number of patients are on antiplatelet therapy for primary and secondary prevention of myocardial infraction or stroke and for prevention of thrombosis after coronary stent placement.^1,2 There are several different classes of antiplatelet agents, which may be used as monotherapy or in combination (Table 1).

Table 1.

Major antiplatelet medications and their properties.

Drug Class	Medication	Half-life	Duration of Action	Clearance	Reversability
Cyclooxygenase-1 inhibitor	Aspirin	15-20 minutes	10 days	Hepatic	Irreversible
ADP receptor inhibitor	Clopidogrel (Plavix)	8 hours	5 days	Hepatic	Irreversible
	Ticagrelor (Brillinta)	7-9 hours	2 days	Hepatic	Reversible
	Prasugrel (Effient)	7 hours	5 days	Hepatic	Irreversible
	Cangrelor (Kengreal)	3-6 minutes	1 hour	Plasma	Reversible
Glycoprotein IIb/IIIa inhibitor	Abciximab (Reopro)	8-12 hours	2 days	Renal	Reversible
	Eptifibatide (Integrilin)	2.5 hours	4 hours	Renal	Reversible
	Tirofiban (Aggrastat)	2 hours	4 hours	Renal, biliary	Reversible
Phosphodiesterase inhibitor	Dipyramidole (Persantine)	24 minutes	3 hours	Hepatic	Reversible
	Cilostazol (Pletal)	11-13 hours	4 days	Hepatic	Reversible

Aspirin irreversibly inhibits platelet function through cyclooxygenase-1 (COX-1) inhibition, resulting in reduced platelet production of thromboxane. Thromboxane is a powerful vasoconstrictor that initiates the platelet release reaction. The half-life of aspirin is 15-20 minutes. However, because it irreversibly inhibits platelet COX-1, its effect persists for the lifespan of the platelets, which is 7-10 days.^3,4

Clopidogrel (Plavix) is a thienopyridine derivative that irreversibly inhibits adenosine diphosphate (ADP) receptor-mediated platelet activation and aggregation. It has a half-life of 8 hours. However, since its effect is irreversible, the anti-platelet effect continues for 7-10 days (the duration to replace the platelet population).² Other ADP receptor inhibitors include ticlopidine (Ticlid), ticagrelor (Brillinta), prasugrel (Effient), and cangrelor (Kengreal). These agents are also known as P2Y₁₂ receptor inhibitors.

Dipyridamole (Persantine) is a pyrimidopyrimidine derivative with antiplatelet and vasodilation properties. It functions as an adenosine reuptake inhibitor. It is used in combination with aspirin and indicated for secondary stroke prevention in patients with cerebrovascular disease.⁵ Dipyridamole has reversible effects on platelet function with an elimination half-life of 10 hours.⁶ Since it is administered as a combination agent with aspirin, therapy would have to be interrupted for 7-10 days to eliminate the antiplatelet effect of both drugs.

Cilostazol (Pletal) is a phosphodiesterase inhibitor with antiplatelet and vasodilatory properties that reversibly inhibit platelet function via cyclic adenosine monophosphate (cAMP) mediated inhibition of platelet activation and aggregation. It can be used in patients with coronary artery disease, typically if there is a coronary stent or peripheral arterial disease.^7,8 Cilostazol has an elimination half-life of 10 hours, and its therapy would need to be interrupted or 2-3 days before surgery to ensure elimination of its antiplatelet effects.⁹

Glycoprotein IIb/IIIa inhibitors are potent platelet inhibitors which block the final common pathway of platelet aggregation. This class includes abciximab (Reopro), eptifibatide (Integrilin), and tirofiban (Aggrastat).¹⁰ Abciximab is a large chimeric monoclonal antibody with a half-life of 8-12 hours, and restoration of normal hemostatic function occurs 3 days after discontinuation. Eptifibatide is a reversible heptapeptide with a half-life of 2.5 hours, and tirofiban is a nonpeptide inhibitor with a half-life of 2 hours. Eptifibatide and tirofiban have recovery of platelet aggregation within 4 hours after completion of infusion.¹¹

Indications for use of antiplatelet therapy

The American Diabetes Association recommends antiplatelet therapy with low dose aspirin for prevention in individuals with diabetes who have coexisting atherosclerotic disease or are at increased risk for cardiovascular disease.¹² The American Heart Association recommends low dose aspirin therapy for all patients with coronary artery disease and combination therapy with aspirin and an ADP receptor inhibitor I patients after acute coronary syndrome (ACS) or percutaneous intervention with stent placement.¹³

A cardiologist should be consulted prior to surgery when patients with coronary artery disease are on dual antiplatelet therapy. The risk of recurrent cardiac ischemia or stent thrombosis is real and non-urgent surgery should be delayed at least 30 days after bare metal stent implantation and optimally 6 months after drug eluting stent implantation. If urgent surgery is needed, and, if the bleeding risk is significant, aspirin should be continued if possible and the ADP receptor inhibitor may be discontinued prior to the procedure and restarted as soon as possible post-operatively.¹⁴

Risk of thrombotic complications from altering antiplatelet therapy

The risk of thrombotic complications has to be weighed against the risk of bleeding when determining whether to continue or hold antiplatelet agents perioperatively. Major thrombotic complications including pulmonary embolism and thrombosis of a prosthetic valve have been reported with discontinuation of antiplatelet therapy perioperatively for cutaneous surgery.¹⁵ These risks are dependent on multiple factors including associated medical conditions, type of surgery, and duration of surgery.^16-18

Risk of bleeding with antiplatelet therapy during ophthalmic surgery

The risk of bleeding while on anti-platelet therapy must be considered when deciding whether to continue therapy during ophthalmic surgery. Bleeding complications include hyphema, vitreous hemorrhage, retinal hemorrhage, choroidal hemorrhage, and retrobulbar hemorrhage.¹⁹ These risks should be weighed against the thromboembolic risks of discontinuing therapy in each individual patient.

Ophthalmic regional anesthesia

The most common types of regional anesthesia used in ophthalmic surgery are topical, peribulbar, and retrobulbar. Of these three approaches, topical anesthesia has the lowest risk of vision threatening hemorrhagic complications, but this approach may not always be an option.²⁰ In a peribulbar block, an anesthetic agent is injected into the subtenon’s space anterior to the equator, blocking CN III and VI and the ciliary nerves. A retrobulbar block involves injection of anesthetic into the intraconal space.^20,21 Retrobulbar hemorrhage is a potential sight-threatening complication of a retrobulbar block.²⁰

A study of 48,862 cataract procedures using data extracted from the Cataract National Dataset compared the complication rates of antiplatelet and anticoagulant medications between users and non-users. Of all the cases, they reported 5.1% were on aspirin, 1.9% were on clopidogrel, and 1.0% were on dipyridamole. They reported an increased minor complication rate secondary to sharp needle or subtenon’s cannula local anesthetic techniques of 8.0% in clopidogrel users compared to 4.3% in non-users. There was no significant increase in potentially sight threatening local anesthetic complication.²²

Prior studies have established the safety of peribulbar and retrobulbar anesthetic blocks for ophthalmic surgery. A prospective study of 16,224 peribulbar blocks demonstrated an incidence rate of 0.74% for orbital hemorrhage, 0.013% for expulsive hemorrhage, 0.006% for globe perforation, 0.006% for grand mal seizure, and no incidences of cardiac or respiratory depression.²³ Another study of 12,500 retrobulbar blocks demonstrated a retrobulbar hemorrhage incidence rate of 0.44% and no statistically difference in visual outcome in these patients compared to controls.²⁴

A systematic review by Takaschima et al investigated the incidence of hemorrhagic complications in patients undergoing needle-based ophthalmic regional anesthesia between patients on antithrombotic therapy and patients not on such therapy. This study included patients on aspirin, clopidogrel, and warfarin in the antithrombotic therapy group. The review reported no difference regarding mild to moderate hemorrhagic complications between the two groups. Additionally, the rates of severe hemorrhagic complications were reported to be very low in both groups as well.²⁰

A retrospective study of 1383 patients undergoing intraocular surgery and receiving anesthesia via a retrobulbar block showed that, of the 482 patients on aspirin, 5% developed lid hemorrhages after stopping aspirin for 0-2 days and 3.2% after stopping aspirin for 3-14 days. The interval from discontinuation of aspirin until surgery did not significantly affect the development of hemorrhage. Additionally, the incidence of hemorrhagic complications in patients receiving aspirin was not statistically different compared to the incidence in patients not receiving antiplatelet therapy (4.1%).²⁵

A prospective study of 115 eyes on aspirin or clopidogrel and undergoing phacoemulsification cataract surgery and receiving a subtenon’s block showed 21% of patients on aspirin developed subconjunctival hemorrhage, 40% on clopidogrel developed a subconjunctival hemorrhage and 19% of controls developed a subconjunctival hemorrhage. No patient developed a sight threatening complication, such as retrobulbar hemorrhage.²⁶

A case-control study of 2,000 patients undergoing intraocular surgery with peribulbar block found no difference in ocular hemorrhage grading or potentially sight threatening complications between patients taking clopidogrel and those who were not.²⁷

These results support the safety of regional needle blocks even in patients on antiplatelet therapy. However, special consideration should be given to patients on dual antiplatelet therapy. A retrospective study of 160,000 patients who received retrobulbar or peribulbar anesthesia revealed that three patients developed a grade IV retrobulbar hematoma, defined as a retrobulbar hematoma with elevated intraocular pressure. Grade IV retrobulbar hematomas are sight threatening and considered an ocular emergency. Of these three cases, one patient was on dual antiplatelet therapy, one was on warfarin monotherapy, and one was on a combination of dual antiplatelet and warfarin therapy.²⁸

Cataract surgery

Numerous studies have been conducted on the use of antiplatelet therapy during cataract surgery. A large prospective cohort study of 19,283 cataract procedures found that the continued use of aspirin did not significantly increase the incidence of hemorrhagic events, including retrobulbar hemorrhage, vitreous hemorrhage, choroidal hemorrhage, and hyphema. Additionally, the study reported the rate of stroke, transient ischemic attack, deep vein thrombosis, myocardial infarction, and myocardial ischemia was similarly low between those who interrupted aspirin use and those who continued aspirin perioperatively.²⁹

A prospective cohort study of patients who underwent cataract surgery found no significant difference in arterial thromboembolic events in 977 patients who interrupted aspirin (0.20%) compared to 3363 patients who continued aspirin (0.65%). Additionally, in these patients, there was no significant difference in the development of ocular hemorrhage between patients who continued (0.06%) or interrupted aspirin therapy (0%).²⁹

These studies support the continued use of antiplatelet therapy during routine cataract surgery. The American College of Chest Physicians (ACCP) has classified cataract and non-cataract eye surgery as low risk for bleeding, which is defined as a less than 2% chance of major bleeding 20 days post-operatively.^30,31 The ACCP recommended continuing aspirin for cataract surgery and interrupting ADP receptor inhibitors for non-high risk cardiac patients while continuing ADP receptor inhibitors for high risk cardiac patients.³⁰ Similarly, the Royal College of Ophthalmology also recommended continuing anti-platelet therapy for ambulatory cataract surgery.³²

Cornea surgery

Suprachoroidal hemorrhage is a rare complication of penetrating keratoplasty, which has also been reported following endothelial keratoplasty as well.^33,34 However, no studies have assessed increased risk of bleeding during corneal surgery in patients on antiplatelet therapy.

Glaucoma surgery

Perioperative use of antiplatelet therapy and elevated preoperative intraocular pressure are potential risk factors for hemorrhagic complications in glaucoma surgery patients. A retrospective study of 367 trabeculectomies, of which 55 were in patients on aspirin, demonstrated a significantly higher rate of hyphema post-operatively, 51% in the aspirin group compared to 28% in the group on no antithrombotic therapy. However, this did not significantly affect intraocular pressure control at two years post-trabeculectomy.³⁵ This study suggests that trabeculectomy is likely to be safe in patients taking aspirin.

Another retrospective study investigated hemorrhagic complications in 347 patients on antiplatelet or anticoagulant therapy prior to glaucoma surgery compared to case-matched controls. Antiplatelet therapy in this study included aspirin, clopidogrel, dipyridamole, or nonselective cyclooxygenase inhibitors. Glaucoma procedures performed in the study included trabeculectomy with and without cataract surgery and tube shunt procedures. Hemorrhagic complications were defined as anterior chamber clot or hyphema requiring intervention, vitreoretinal hemorrhages, and choroidal hemorrhages. The study found that patients who were continued on antiplatelet therapy had a 6.9% rate of hemorrhagic complications. Patients on antiplatelet therapy who discontinued antiplatelet therapy for surgery had a 13.5% hemorrhagic complication rate. There was no statistical significance between the group that continued antiplatelet therapy and the group that discontinued therapy perioperatively. Among case-matched controls, 3.7% developed hemorrhagic complications, which was statistically significantly lower than both the continued and discontinued antiplatelet therapy groups.³⁶

There is limited data on the complication rate with perioperative antiplatelet therapy in glaucoma surgery. A consensus report by the Italian cardiological, surgical, and anesthesiological societies considered trabeculectomy to be an intermediate risk procedure for hemorrhagic complications and recommended that aspirin may be continued, but ADP-receptor inhibitors should be interrupted 5 days prior to surgery and resumed within 24-72 hours. For high thrombotic risk individuals, they recommended bridge therapy with GP IIb/IIIa inhibitors.³⁷ However, further research is needed to discern the safety of these agents.

Strabismus surgery

There is limited published data on the use of antiplatelet therapy perioperatively for extraocular muscle surgery.

Vitreoretinal surgery

A number of studies have been reported in recent years on the hemorrhagic complication rate associated with antithrombotic agents in vitreoretinal surgery. A prospective study of 541 eyes undergoing vitreoretinal surgery, including 60 of patients on aspirin, demonstrated choroidal hemorrhage in 2% of cases. Of these, none were on antiplatelet therapy with aspirin. This study suggested no increase in the rate of hemorrhagic complications in patients taking aspirin and undergoing vitreoretinal surgery and recommended that aspirin therapy be continued perioperatively.³⁸

Mason et al studied hemorrhagic complications of warfarin and clopidogrel in 289 patients undergoing 25-guage vitrectomy surgery. There was no significant difference in the rate of transient vitreous hemorrhage between the warfarin, clopidogrel, and control group. No patient experienced hemorrhagic complications from peribulbar or retrobulbar anesthetic block. No choroidal or retrobulbar hemorrhages occurred in any patient. The study concluded that patients should continue their therapeutic antithrombotic regimen without interruption.³⁹

An observational retrospective case-control study of 822 patients undergoing vitreoretinal surgery found that there was no statistically significant difference in the incidence of early postoperative intraocular hemorrhage in patients on antiplatelet therapy and no reoperation or surgery failure was attributable to antiplatelet therapy. The authors recommended that antiplatelet agents should be suspended perioperatively whenever safely possible to reduce the risk of hemorrhagic complications.⁴⁰

A prospective study of 107 vitreoretinal procedures in patients taking aspirin, clopidogrel, or warfarin found that the single greater independent predictor of intraoperative hemorrhage was proliferative diabetic retinopathy. Additionally, the single greatest independent predictor of postoperative bleeding was the presence of diabetes mellitus.⁴¹

Recently, Meillon et al published a prospective multicenter study of 804 vitreoretinal procedures compared hemorrhagic complications between patients treated or not treated with antiplatelet or anticoagulant agents. Among these procedures, 18.4% were treated with antiplatelet agents (aspirin, clopidogrel, or aspirin with clopidogrel), 96.5% of which were continued on antiplatelet therapy perioperatively. The study concluded that antiplatelet therapy was not a factor associated with the development of hemorrhagic complications.⁴²

A case-control study of 322 patients undergoing primary rhegmatogenous retinal detachment repair while being treated with aspirin found that aspirin was not significantly associated with hemorrhagic complications during or after surgery. Scleral buckling was not significantly associated with hemorrhagic complications amongst these patients. Pars plana vitrectomy did have a statistically significant association with hemorrhagic complications, but confounding factors could not be excluded.⁴³

Based upon the conclusions of these studies, antiplatelet therapy can likely be continued safely in the perioperative period of vitreoretinal surgery, especially in surgeries at low risk of hemorrhage. Potential discontinuation of antiplatelet therapy should be reserved for patients at high risk of bleeding during surgery in collaboration with the anesthesiologist and cardiologist.

Oculoplastic and lacrimal surgery

Intraoperative or postoperative bleeding in oculoplastic and lacrimal surgery may lead to vision loss or permanent functional deficits.^44,45 Many variables affect the likelihood of hemorrhage during surgery, including inadequate cauterization and poor surgical technique. Visualization is of utmost important during endonasal procedures, and obstruction secondary to hemorrhage can lead to poor outcomes. When the view is obscured due to persistent bleeding, the risk of graft failure also increases due to hematoma formation.⁴⁶

A retrospective chart review of 1130 oculoplastic surgical procedures was performed to evaluate bleeding complications in patients continuing antithrombotic therapy, interrupting antithrombotic therapy perioperatively, and those not on antithrombotic therapy. One of the 682 cases not on thrombotic therapy (0.15%) developed a post-operative orbital hematoma. Three of the 207 patients who interrupted antithrombotic therapy (1.45%) developed increased bleeding intraoperatively. Of these three cases, two patients were previously on aspirin and one was previously on warfarin, an anticoagulant. Of this same cohort, there was one case (0.48%) of postoperative hematoma after lower lid blepharoplasty in a patient interrupting aspirin therapy. Of the 145 patients who continued antithrombotic therapy, there was one case (0.69%) of postoperative bleeding requiring wound exploration and cauterization in a patient who had been continued on aspirin and held clopidogrel prior to surgery. None of the patients in the study had permanent visual deficits or deformity secondary to hemorrhage.⁴⁷

One prospective study of 1575 oculoplastic procedures was performed. Of these, 471 procedures were performed on patients who used aspirin, dipyridamole, or ticlopidine regularly. Antiplatelet therapy was discontinued preoperatively in 52% of these patients. The results of this study demonstrated no significant difference between patients who continued antiplatelet therapy and those who did not. Of note, in this study antiplatelet therapy was discontinued in patients undergoing dacryocystorhinostomy (DCR), enucleation, or extensive orbital surgery if they did not have a history of stroke or ACS.⁴⁴

A study of 9204 facial plastic surgical procedures comparing complication rates in patients on antiplatelet or anticoagulant therapy to controls showed that patients receiving aspirin at the time of surgery were not more likely to have a complication compared to controls. The study concluded that patients undergoing facial plastic surgery may continue taking antiplatelet therapy during the perioperative period safely with minimal serious complications.⁴⁸

These results suggest that many patients can safely undergo oculoplastic surgery without stopping antiplatelet therapy, but the decision to withhold or continue these agents should be individualized based upon the patient’s medical history and the type of procedure to be performed. Despite data supporting low rates of complications during oculoplastic and facial plastic surgery, many leaders in the field still recommend discontinuation of antiplatelet therapy prior to procedures considered to be higher risk for vision threatening complications, including blepharoplasty, lacrimal surgery, and deep orbital surgery.^44,47

Reversal of antiplatelet therapy

There is no pharmacologic agent that can reverse the antiplatelet effects of agents that irreversibly inhibit platelet function, such as aspirin, clopidogrel, or ticlopidine. Patients who require urgent surgical procedures and necessitate normalization of their platelet function can be transfused with platelets.⁴⁹ Prohemostatic agents can be used as an alternative to platelet transfusion in individuals who have been exposed to antiplatelet drugs. These agents include ε-aminocaproic acid, tranexamic acid, and 1-deamino-8-D-arginine vasopressin.⁵⁰ However, due to the lack of evidence on the safety and efficacy of these reversal agents to neutralize antiplatelet effects of drugs, such as aspirin and clopidogrel, it is reasonable to limit their use to those who are at risk of excessive or life-threatening bleeding in the perioperative period.

Management of the operative field

Meticulous surgical technique and an understanding of the anatomy can help to minimize surgical trauma and decrease the risk of hemorrhage. Use of hypotensive anesthesia can also reduce intra-operative bleeding.^51,52 Local epinephrine can assist in achieving hemostasis, especially in oculoplastic and lacrimal surgery.⁵³ Radiofrequency cutting diathermy can minimize bleeding at the operative site by cauterizing the wound at the time of incision.⁵⁴ Carbon dioxide (CO₂) lasers can be used to minimize bleeding by a similar mechanism.⁵⁵

Fibrin sealants can be used to improve hemostasis at the operative site.^56,57 They function by duplicating the last stage of blood clot formation. A combination of thrombin, fibrinogen, ionized calcium, and factor XIII are applied at the site where hemostasis is desired. Thrombin converts fibrinogen to fibrin and activates factor XIII, resulting in cross-linking of fibrin and strengthening of the clot.⁵⁸ Hemostatic sponges can also be used to minimize blood loss.⁵⁹

Conclusions

Antiplatelet use does not appear to be associated with significant bleeding in patients undergoing most types of ophthalmic surgery or infiltration of local anesthesia. In procedures where hemorrhagic risk may be higher or unknown or in cases where persistent bleeding from the operative site may result in poor surgical outcomes, antiplatelet therapy may be discontinued preoperatively in patients without significant cardiovascular risk factors. It is important to consult with both anesthesiology and cardiology specialists involved in the patient care prior to discontinuation of antiplatelet medications.