Anticoagulants and Antiplatelet Agents
Anticoagulants and antiplatelet agents pose a great risk in patients undergoing surgery. Because these products inhibit at least one, and often multiple, factors in the clotting cascade, patients have an increased risk of bleeding with invasive procedures. Patients also have an increased risk of developing a thromboembolism if these drugs are withheld for any length of time. Health care professionals must consider the following factors when considering perioperative use of these drugs: type of surgical procedure, type of anesthesia, patient-specific risk factors, and anticoagulation agent being used.
Patient-specific risk factors include the medical indication for anticoagulation, as well as any individual risk factors that patients may have for developing a venous thromboembolism (VTE). Indications for anticoagulation include deep vein thrombosis (DVT), pulmonary embolism (PE), acute coronary syndrome, and prophylaxis for mechanical heart valves. The most significant risk factors include advanced age; prolonged immobility, stroke, or paralysis; previous VTE; cancer; major surgery; trauma; obesity; varicose veins; cardiac dysfunction; indwelling central venous catheter; inflammatory bowel disease; nephrotic syndrome; and pregnancy or estrogen use.
Each type of surgical procedure requires different precautions. For all procedures, clinical judgment should be exercised when weighing the risks and benefits of anticoagulation-related bleeding complications versus the development of a thromboembolism. There are three specific categories of surgical procedures: procedures that carry a minimal bleeding risk, procedures with an increased risk of hemorrhage, and those in between. Some of the more significant types of surgical procedures are discussed below.
The nature of neurosurgical procedures complicates the issue of anticoagulation. The major issues surrounding the use of these agents are when to restart anticoagulation after the procedure and how to determine which patients have a greater risk of thromboembolic complications if these medications are stopped. A retrospective review of 39 patients with intracranial hemorrhage (ICH) and mechanical heart valves was conducted to evaluate these issues. Of the 26 patients who survived the ICH, one third had prothrombin times above their target range. All 26 were treated with fresh frozen plasma and vitamin K, and 14 had a craniotomy and evacuation of the hematoma. Although anticoagulation was discontinued for a median of eight days (range, two days to three months), no systemic or cerebral embolic or hemorrhagic complications occurred. The authors concluded that discontinuation of anticoagulation for one to two weeks should be sufficient for most patients in order for health care providers to observe the evolution of a lobar or ganglionic hemorrhage, clip or coil an aneurysm, or evacuate an acute subdural hematoma.
To assess when to restart anticoagulation, one study retrospectively reviewed the charts of 27 patients with a warfarin-related ICH. In all patients, warfarin was discontinued when ICH was diagnosed, and all patients received vitamin K. While 17 patients had surgery, 10 were treated conservatively. Of the 17 who had surgery, 9 restarted warfarin therapy within three days of the procedure, and none of these 9 had intracranial rebleeding.
A recent review offered several strategies for anticoagulation in neurosurgical procedures. The authors recommended that oral anticoagulants should be stopped two to three days before elective surgery and, in emergent cases, vitamin K or fresh frozen plasma should be administered until near-normal coagulation is achieved. They also recommended that oral anticoagulation should not be restarted in the first 24 to 48 hours after surgery and developed recommendations based on thromboembolic risk status. High-risk patients (onset of DVT or PE within the three months before surgery) should receive a vena cava filter and subcutaneous heparin, and oral anticoagulation should be restarted on postoperative days 3-5. Moderate-risk patients (e.g., those with mechanical valves, valvular disease with atrial fibrillation, or carotid dissection) should receive subcutaneous heparin postoperatively, and oral anticoagulation should be restarted on postoperative days 5-7. Low-risk patients (e.g., those with isolated atrial fibrillation, aortic valve prosthesis, old DVT or PE) should receive subcutaneous heparin postoperatively, and oral anticoagulation should be restarted on postoperative days 7-14.
The American Society for Gastrointestinal Endoscopy recommends continuing oral anticoagulation during low-risk gastrointestinal procedures, such as diagnostic esophagogastroduodenoscopy, diagnostic endoscopic retrograde cholangiopancreatography, flexible sigmoidoscopy, and colonoscopy. For high-risk procedures, such as colonoscopic or gastric polypectomy or other procedures with a high potential for bleeding, oral anticoagulants should be discontinued for three to five days before the procedure. For patients undergoing high-risk procedures who are also considered at high risk of thromboembolism (e.g., atrial fibrillation associated with valvular heart disease or mechanical valves, prior thromboembolic event), bridge therapy with heparin should be considered.
Several studies have shown a low risk of clinically important perioperative bleeding when oral anticoagulants were continued during cataract surgery. A case series of 13 patients undergoing cataract extraction while taking warfarin showed no increase in procedural or postoperative bleeding. A prospective evaluation assessed the need for discontinuing anticoagulation in 2178 cataract surgeries. In the 28 patients in whom anticoagulation was continued, an increase in lid ecchymosis and subconjuctival hemorrhage was noted. The authors deemed this clinically insignificant, and concluded that the risk of serious life-threatening complications from stopping these agents outweighs the possibility of sight-threatening bleeding complications. A retrospective study specifically assessed the effect of continuing or discontinuing anticoagulation on the frequency of hemorrhagic and thrombotic complications of 50 patients undergoing ocular surgery. In 9 of the patients, warfarin therapy was continued, and warfarin was discontinued an average of 5.5 days (range, 3-9 days) before the procedure in the other 41 patients. While there was an increased frequency of hemorrhagic complications in the warfarin patients compared with historically matched controls (p < 0.03), there was no difference in hemorrhagic complications between these groups. Therefore, it is generally recommended that oral anticoagulants be continued during most ophthalmologic procedures.
Studies have also failed to show an increase in hemorrhagic complications during dermatological surgery, and the current recommendation is for oral anticoagulants to be continued during these procedures.[67,68,69] A prospective study examined the rate of postoperative bleeding in patients taking aspirin (n = 81), warfarin (n = 12), or nonsteroidal anti-inflammatory drugs (NSAIDs) (n = 16) undergoing Mohs' surgery compared with a control group (patients undergoing the same procedure not taking these agents) (n = 213). Of the 322 patients enrolled, excessive intraoperative bleeding was seen in 5 patients (41.7%) taking warfarin (p = 0.0003), 1 patient (6.2%) taking NSAIDs, 9 patients (11%) taking aspirin (p = 0.052), and 9 control patients (4.2%). The bleeding was easily controlled, and there was no difference in postoperative bleeding between groups. A retrospective, controlled study of 653 patients undergoing excisional and Mohs' surgery showed no difference in severe complications for patients who were taking warfarin or platelet inhibitors compared with historical controls (1.6% versus 0.7%, respectively). Of particular importance, the frequency of severe complications was not increased, regardless of whether these medications were suspended before the procedure. A similar study showed no increased risk of perioperative bleeding during cutaneous surgery in 16 patients taking long-term warfarin therapy, compared with 71 controls. In addition to the low risk of bleeding, two reports found an increased rate of significant thromboembolic complications, such as stroke and PE, associated with the discontinuation of oral anticoagulants before surgery.[70,71] The authors of both of these reports concluded that, because of this increased risk, oral anticoagulants should not be withheld before dermatological surgeries.
Hemorrhage associated with the use of anticoagulants may not always be related exclusively to the surgical procedure. Preoperative anticoagulation can also be problematic, depending on the type of anesthesia used. General anesthesia typically does not carry any heightened risk, but regional nerve block may be considered a relative contraindication. In 1997 a Food and Drug Administration public health advisory reported that 43 patients who had received spinal or epidural anesthesia with concomitant enoxaparin developed a spinal hematoma. To reduce the risk of bleeding during spinal or epidural anesthesia, guidelines have been developed to help improve the safety of patients who have already received, or will be receiving, prophylactic anticoagulation ( Table 2 ).
The current treatment options available comprise warfarin, unfractionated heparin (UFH), and low-molecular-weight heparin (LMWH). Each of these agents is associated with independent risks and benefits and should be managed differently at the time of surgery.
Warfarin is used as a short-term treatment for acute thrombotic events and for long-term prophylaxis of thromboembolism in high-risk patients. As a result, the majority of patients on warfarin who require an invasive procedure are often from the ambulatory care setting. As discussed previously, patients may or may not be required to cease warfarin therapy, depending on the type of surgery. However, if the risk of VTE outweighs the risk of a potential bleeding episode (e.g., high-risk patients, such as those with mechanical valves), bridge therapy with UFH or an LMWH may be used. The typical outline provided in much of the literature suggests the following: reverse the oral anticoagulation, bridge with UFH or LMWH until surgery, resume the UFH or LMWH postoperatively (along with warfarin), and discontinue UFH or LMWH when warfarin reaches a therapeutic level.
Bridge therapy with LMWHs has been evaluated in several studies.[74,75,76] The first, a prospective cohort study, examined 112 patients on long-term warfarin therapy undergoing various surgical procedures. Warfarin was discontinued at least five days before the procedure, and patients were given s.c. dalteparin 100 units per kilogram twice daily five days before surgery. Patients were restarted on dalteparin 8-12 hours postoperatively, along with their standard warfarin dosage. None of the patients in this study had a thromboembolic event, although 1 patient had bleeding at the injection site.
A prospective cohort study of 21 patients undergoing major surgery or an invasive procedure receiving long-term warfarin therapy was conducted. Warfarin therapy was stopped five days before surgery for patients with an international normalized ratio (INR) of 2-3, and six days before if their INR was 3-4.5. Bridge therapy with s.c. enoxaparin 1 mg per kilogram twice daily was started 36 hours after discontinuing the warfarin. The enoxaparin was discontinued 12-18 hours before surgery and was resumed, along with the patient's standard warfarin dosage, an average of 13.5 hours after surgery until the INR was in the therapeutic range. None of the patients developed bleeding or thrombotic complications during the procedure, although 3 patients developed soft tissue bleeding postoperatively. No thromboembolic events occurred during the postoperative period.
Another study evaluated 24 patients undergoing elective noncardiac or invasive procedures. Warfarin was discontinued four days before the procedure, and patients were given s.c. dalteparin 200 units per kilogram every morning for the two days before surgery. If the patient had an INR of >1.6 on the day before surgery, 1 mg of p.o. vitamin K was given. On the morning of surgery, no dalteparin was given, and warfarin was restarted that evening. Dalteparin was restarted on postoperative day 1 and continued until the patient achieved an INR of >1.9. There were no major bleeding complications during the study. One patient had a thromboembolic complication (transient ischemic attack) postoperatively, which may have been due to a subtherapeutic INR.
If reversal of warfarin is necessary preoperatively, vitamin K may be administered, but must be given 24-48 hours before the procedure. If vitamin K is chosen, the oral route may be preferred because it yields a more rapid and predictable response than does subcutaneous administration. In addition, the oral route eliminates the risk of anaphylaxis, which may occur with intravenous administration. If emergent reversal is required, fresh frozen plasma is the preferred agent.
The lack of controlled trials in this area makes it difficult to formulate recommendations with absolute certainty. When normal coagulation is required preoperatively, warfarin should be suspended for five days before the procedure to allow the INR to fall below 1.5. Bridge therapy with UFH or LMWHs may be necessary for high-risk patients when the warfarin is withdrawn before the procedure.
Full-dose i.v. UFH has been the agent of choice for patients requiring anticoagulation before surgery. The advantage of UFH is its short elimination half-life (1 to 2 hours). Its disadvantages are the requirements for hospitalization and continuous monitoring of the activated partial thromboplastin time (aPTT). If the aPTT is in the target range, the heparin infusion should be stopped 6 hours before the surgical procedure. If additional reversal is required, the use of protamine may be considered. Restarting the UFH infusion should be delayed for 12 hours after the procedure or longer if there is evidence of bleeding from the surgical site, and no bolus dose should be given.
If a patient is receiving thromboembolic prophylaxis via subcutaneous heparin, the drug should be discontinued for 12 hours (or one dose should be skipped) before surgery because absorption of subcutaneous heparin is slowed, which delays elimination of the drug. The reinstitution of subcutaneous UFH should be avoided for at least 12 hours postoperatively or longer if signs of bleeding are evident.
As an alternative to heparin, LMWHs are a viable option for both the treatment and prophylaxis of thromboembolism. These drugs are well tolerated and easy to administer in the outpatient setting; therefore, hospitalization required for the bridging of anticoagulation therapy may be avoided. While LMWHs can be used in high-risk patients, including those with prosthetic valves, newer data suggest avoiding their use in pregnant patients with prosthetic valves. Because the half-life of LMWHs is longer than that of UFH and the subcutaneous absorption is delayed, LMWHs should be held for at least 12 hours before surgery.[73,78] The postoperative reinstitution of LMWHs should follow the same recommendations as those for UFH.
Antiplatelet products, such as aspirin and the thienopyridines (clopidogrel and ticlopidine) are commonly used in patients for indications that often require, or are related to, surgical procedures. These agents irreversibly inhibit platelet aggregation by different mechanisms. Aspirin inhibits the cyclooxygenase (COX) activity of prostaglandin-H synthase, which leads to the inhibition of thromboxane A2 (TXA2) and prostaglandin I2, both of which are responsible for platelet aggregation. Ticlopidine and clopidogrel appear to act through the adenosine diphosphate (ADP) receptor and may reduce the number of ADP-binding sites; however, the exact mechanism for their platelet-inhibitory effects is unknown. Since these effects last for the lifetime of the platelet (approximately 10 days), it takes at least 7 days for platelet function to return to 70% of normal after the administration of these agents. Several studies have described an increased frequency of postoperative bleeding associated with the use of antiplatelet agents,[84,85,86,87,88,89,90,91] particularly with CABG surgery, but the clinical significance of CABG-related bleeding has been questioned.[10,92] It is routinely recommended to discontinue aspirin, clopidogrel, and ticlopidine at least 5-7 days before surgery to reduce the risk of perioperative bleeding and reinstitute them when the bleeding risk has diminished.
NSAIDs are commonly used by patients undergoing elective surgical procedures. The antiinflammatory effects and analgesic effects of these medications are caused by the inhibition of prostaglandins, which also affects platelet aggregation. NSAIDs can inhibit TXA2-dependent platelet function through competitive, reversible inhibition of COX-1 as long as the NSAID remains in the bloodstream; therefore, the half-life of NSAIDs is a major contributor to their potential effect on platelet function and the time required to suspend these medications before surgery. Short-acting NSAIDs should be stopped at least one day before surgery, while longer acting agents should be discontinued two to three days in advance.[77,93] Since platelets lack COX-2, specific inhibitors of this enzyme, such as celecoxib, have a minimal effect on platelet aggregation but should be withdrawn for two to three days before surgery because of their effects on renal function. Antiplatelet and NSAID therapy can be resumed soon after surgery if patients are stable and have an adequate volume status. A summary of the pertinent information for these agents is provided in Table 3 .
Am J Health Syst Pharm. 2004;61(9) © 2004 American Society of Health-System Pharmacists
Cite this: Discontinuation and Reinstitution of Medications During the Perioperative Period - Medscape - May 01, 2004.