Urological Surgery and Antiplatelet Drugs After Cardiac and Cerebrovascular Accidents

Antiplatelet Agents 

Antiplatelet agents are classified into the 3 categories of acetylsalicylic acid (aspirin), thienopyridines (clopidogrel, prasugrel) and platelet GP-IIb-IIIa inhibitors (eptifibatide, tirofiban, abciximab). This review deals mainly with aspirin and clopidogrel since to our knowledge there is no perioperative experience with prasugrel, and GP-IIb-IIIa inhibitors are only used for ACS and immediately after PCI.

Aspirin irreversibly inhibits thromboxane A2 and prostacyclin (PGI2) synthesis. It is effective in doses of 50 to 160 mg a day. There is no evidence that doses greater than 160 mg a day are more efficacious in reducing the cardiovascular risk but they increase the risk of gastric bleeding. After cessation of aspirin, platelet aggregation returns to baseline in 5 days. The long-term benefit of aspirin is a RRR of recurrent vascular events of 38% after MI and 25% after stroke.3

Clopidogrel is a pro-drug oxidized by hepatic cytochromes into an active metabolite which has a plasma elimination half-life of 8 hours and inhibits irreversibly platelet aggregation in a dose dependent fashion (daily dose 75 mg). At 7 days after cessation 80% of platelets have recovered normal aggregation. There are no major differences in terms of bleeding between aspirin and clopidogrel when administered alone. The benefit of dual therapy (aspirin and clopidogrel) compared to aspirin alone has been clearly demonstrated. The 1-year RRR is 31% after PCI and stent.4 Beyond 1 year the benefit of dual therapy is less pronounced (RRR 23%) but it is better than aspirin alone after PCI and DES. The rate of MI and death is 3.1% in patients still on clopidogrel at 2 years vs 7.2% in those who discontinue clopidogrel at 6 months.5

Evidence Acquisition 

A PubMed search was conducted for articles published during the last 10 years using the key words antiplatelet agents withdrawal, perioperative antiplatelet agents and coronary stents, perioperative antiplatelet agents after cerebrovascular accidents, antiplatelet agents and management of surgical bleeding, antiplatelet agents substitution and surgery. A total of 778 articles were found and 167 matching the focus of this review were selected, including controlled trials, observational series, meta-analyses and guidelines from major medical associations.

Evidence indicates that early discontinuation of antiplatelet therapy is associated with a worse outcome in patients after cardiac or cerebrovascular accidents.5, 6, 7, 8, 9 However, most of the data supporting maintenance of antiplatelet therapy in the perioperative period are from case series and nonrandomized retrospective or prospective studies because of the obvious ethical difficulties in performing a placebo controlled, randomized trial when the tested substance is possibly a matter of safety. The recent recommendations formulated by panels of experts are essentially based on precautionary principles and an empirical balance between the risk of vessel thrombosis when antiplatelet agents are stopped and the risk of surgical hemorrhage when they are maintained.10, 11, 12 Our review demonstrates whether the risk of coronary thrombosis outweighs the risk of surgical hemorrhage when antiplatelet agents are withdrawn before surgery.

Antiplatelet Agents and Coronary Revascularization 

As long as coronary stents are not fully covered by a cellular layer they behave like unstable plaques and require dual antiplatelet therapy. The metal frame of BMSs is covered by smooth muscle cells within 6 weeks and by a normal endothelial layer within 3 months but only 13% of DESs are completely covered by endothelium at 3 months and no more than 56% are covered at 3 years.13

Several retrospective series demonstrated high complication rates when surgery was performed early after PCI.14, 15, 16 The combined rate of MI and death is 10% to 38% within 4 weeks after BMS but decreases to 3.8% and 2.8% when surgery is performed at 2 and 3 months, respectively (fig. 1). After DES the rate of adverse cardiac events is more continuous (5.9% up to 12 months) and decreases to 3.3% beyond 1 year but the mortality rate is high (average 35%).17

View Large Image Download to PowerPoint

Figure 1. Incidence of major adverse cardiac events (MACE) according to delay (months) between stent implantation and noncardiac surgery. Events include MI, stent thrombosis, target vessel revascularization and death.16, 17

BMSs are threatened by an overgrowth of the neoendothelium, which leads to a restenosis rate of 12% to 25% at 6 to 12 months. To prevent this phenomenon DESs slowly elute antiproliferative agents. The rate of restenosis decreases to 6.5% at 4 years but the slow re-endothelialization necessitates longer dual antiplatelet therapy. The recommended duration of clopidogrel treatment of 4 to 6 weeks after BMS must be extended to at least 12 months after DES according to recent guidelines10, 11, 12 based on the rate of re-endothelialization13 and clinical trials.5, 6 This duration should be considered as a minimal safety measure. Although rare (0.6% yearly increase compared to BMS), late DES thrombosis is a catastrophic event with a high mortality rate (19% to 45%) because it corresponds to the acute interruption of flow in a previously normal throughput vessel. In all studies the major independent predictor of stent thrombosis is antiplatelet therapy cessation.

Antiplatelet Agents and Cerebrovascular Disease 

The combination of aspirin and clopidogrel is no more efficacious than the use of each medication alone to prevent recurrent stroke, MI or death but it increases the risk of bleeding (2.6% vs 1.3%).18 The addition of extended release dipyridamole (400 mg) to aspirin (50 to 150 mg) has a RRR of 20% compared to aspirin alone without increasing the risk of hemorrhage.19 Whereas aggressive antiplatelet therapy is beneficial in coronary artery disease, when most acute events are caused by unstable plaque rupture, intense platelet inhibition seems to be not efficacious for secondary stroke protection when the incidence of unstable atheromatous plaque is smaller and the rate of major bleeding, particularly intracranial, is dangerously increased.18

Evidence for Thrombotic Risks at Withdrawal of Antiplatelet Agents 

Antiplatelet Discontinuation in the Surgical Setting 

The situation is even more risky in the perioperative period, which is characterized by increased platelet aggregability and decreased fibrinolysis. Early interruption of dual antiplatelet therapy to allow major noncardiac surgery during the first month after PCI with BMS leads to a cardiac mortality of up to 86% vs 5% when the treatment is maintained perioperatively.14 In an observational study of 192 patients the incidence of DES thrombosis after early noncardiac surgery (13% urological operations) was 31% in those who stopped taking clopidogrel before the required delay vs 0% in those who continued dual antiplatelet therapy.20 All patients with stent thrombosis died. Pooling the data from different observational studies of noncardiac surgery in patients with coronary stents revealed that mortality is directly related to the delay between coronary revascularization and noncardiac surgery in patients taken off antiplatelet therapy before surgery (fig. 2). The shorter the delay, the higher the mortality rate. Maintaining antiplatelet therapy dramatically decreases the mortality rate from 85% to 5% and 31% to 0%, respectively.14, 20

View Large Image Download to PowerPoint

Figure 2. Mortality of patients with coronary stents taken off AP drugs for noncardiac surgery according to delay since coronary revascularization (Artang R: Am J Cardiol 2007; 99: 1039; Brotman DJ: J Hosp Med 2007; 2: 378; Rhee SJ: Int Heart J 2008; 49: 135; Kim HL: Int J Cardiol 2008; 123: 353). Two studies have comparative data on mortality of patients operated on without cessation of AP drugs (5% and 0%, respectively).14, 20

Evidence for Surgical Hemorrhage With Antiplatelet Therapy 

The 3 best investigated interventions regarding antiplatelet therapy are transrectal P-Bx, TURP and ureteroscopy. A study of 1,810 patients after P-Bx showed no statistical increase in bleeding complications when the aspirin regimen was maintained (3.7% vs 2.5%).21 Two prospective, randomized trials revealed no significant difference in the incidence of hematuria, hematospermia or rectal bleeding in patients maintained on aspirin during P-Bx, although the duration of bleeding was significantly prolonged.22, 23 A study of 387 patients demonstrated an increase in minor bleeding complications with a higher incidence and duration of hematuria and rectorrhagia but no difference in severe bleeding.24

Flexible ureteroscopy with stone disintegration (holmium:YAG laser) appears to be safe in patients on clopidogrel or aspirin.25 None of the procedures had to be terminated due to poor visibility from bleeding. There was no difference in the stone-free rate, intraoperative and postoperative complications or hemorrhagic adverse events. Although the hemoglobin decrease was greater when antiplatelet agents were maintained (0.6 vs 0.2 gm/dl, p <0.0001), the difference was not clinically relevant.

For conventional TURP the relationship between increased postoperative hemorrhage and use of aspirin was first documented in 1993,26 and emphasized by a retrospective analysis showing that more blood units were administered to patients taking aspirin.27 Therefore, these authors recommended the withdrawal of nonsteroidal anti-inflammatory drugs 1 week before TURP. This recommendation was challenged by studies showing no significant increase in blood loss if patients were taking aspirin.28, 29 However, there was a higher incidence of late bleeding with tamponade of the bladder requiring emptying in the operating room.29 In a randomized controlled study of patients taking 150 mg aspirin there was no difference in intraoperative hemorrhage, operation time and amount of tissue resected but there was an increase in postoperative blood loss of 51% (284 vs 144 ml).30 There was no difference in time to catheter removal and hospital stay. The number of patients requiring blood transfusion was higher in the aspirin group, although only 3 (11%) received 3 or more units of blood. Early reinitiation of aspirin if stopped 5 days before TURP was investigated in a cohort of 120 patients.31 Starting aspirin 24 hours after discontinuation of bladder irrigation or 3 weeks after surgery did not influence the time to catheter removal or persistent hematuria.

Holmium laser enucleation and vaporization of the prostate (potassium titanyl phosphate laser) are new methods of TURP that reduce bleeding in patients receiving antiplatelet therapy.31, 32 Published reports show no significant increase in postoperative complications if aspirin and/or clopidogrel is continued throughout surgery. Furthermore, the long-term functional results are comparable. In summary, there is evidence that P-Bx, flexible nephroscopy and TURP can be performed in patients receiving antiplatelet agents.

The fact that many urological procedures have not been investigated prevents making meaningful recommendations for larger urological surgeries including nephrectomy, cystectomy or prostatectomy. For radical prostatectomy the technique of robotic assistance, combining the increased abdominal pressure with a more intuitive handling of surgical instruments, shows a trend toward less blood loss during surgery.33 The average blood loss in patients not on antiplatelet therapy undergoing radical prostatectomy with robotic assistance is 100 to 200 ml34 compared to 800 ml for conventional radical prostatectomy.35 Whether this holds true for patients on antiplatelet therapy remains to be studied.

When considering other noncardiac surgeries to define a strategy for urological patients, we realized that large randomized trials comparing the effects of antiplatelet agents with placebo in surgical settings are also lacking. Most data are provided by retrospective or observational studies, usually with low level evidence, and are underpowered to answer the question of the increase in operative morbidity and mortality due to aspirin or dual antiplatelet therapy. A meta-analysis of nonurological surgeries including 474 studies comparing surgical bleeding of patients operated on with or without aspirin, revealed an average intraoperative hemorrhagic risk increase of 1.5-fold with aspirin but no change in the mortality and complication rates.36 The transfusion rate increased by 20% for some orthopedic operations.

Surgical data comparing aspirin and clopidogrel alone are lacking but in nonsurgical trials the rate of spontaneous hemorrhage was similar with both drugs.4 Although dual antiplatelet therapy is associated with an average 50% increase in surgical hemorrhage, the clinical evidence is limited to some observational series in vascular,37 visceral38 and transbronchial surgery.39 There was an increase in oozing and diffuse bleeding but no increase in major blood loss for major nonurological surgeries.16, 20 There was no clear increase in morbidity and mortality except during intracranial neurosurgery. In 3 studies the rate of perioperative transfusion during nonurological surgeries for patients with and without dual antiplatelet therapy was not statistically different, although there was a slight trend toward an increased rate in those receiving dual antiplatelet therapy (43% vs 38%,15 24% vs 20%20 and 15.4% vs 15.0%,17 respectively).

Evidence Synthesis 

The available reports indicate that the increase in blood loss related to perioperative continuation of aspirin and clopidogrel does not appear to cause an increase in surgical complications or mortality except when bleeding occurs in a closed space like the skull, spinal canal or posterior chamber of the eye, and when surgery is accompanied by massive hemorrhage and difficult hemostasis.40 Many urological operations must be placed in the difficult hemostasis group, including partial nephrectomy, prostate surgery and cystectomy. From our survey of the current literature we estimate that the average increase in bleeding during noncardiac surgery is approximately 20% with aspirin and up to 50% with dual antiplatelet therapy, and the transfusion rate is inconsistently affected.15, 17, 20 The global disadvantage of requiring transfusion (short-term complications 0.4%,41 long-term survival reduced by 16%42) is far less than the 35% average mortality rate when the drugs are withdrawn before surgery. Nevertheless, blood transfusions should be avoided when possible by treating preoperative anemia, and resorting to a highly trained surgical team familiar with modern blood sparing techniques and strategies.

The risk of MI and mortality in patients operated on while receiving adequate antiplatelet therapy should be the same as that with stable coronary artery disease, that is a MI rate of 2% to 6% and a mortality rate of 1% to 5% depending on the surgical procedure.40 However, the risk of withdrawing antiplatelet therapy is associated with a MI rate of 20% to 40% and a mortality rate of 19% to 85% depending on the delay between revascularization and surgery. Therefore, the risk of coronary thrombosis appears higher than that of surgical hemorrhage. Pending high level of evidence data on the risk-benefit balance of stopping vs continuing antiplatelet therapy, preoperative cessation of aspirin and/or clopidogrel should be avoided when possible.

Current Recommendations and Proposals 

It is challenging to define clear-cut recommendations for urological patients because this population has not been sufficiently studied. However, the most recently published guidelines have concluded that premature discontinuation of antiplatelet therapy before surgery is dangerous due to the dramatic increase in MI and death.10, 11, 12

Aspirin has recently been recommended as a lifelong therapy which should be continued perioperatively when prescribed for secondary prevention after ACS, MI, stroke, vascular surgery or PCI with any type of stent, regardless of the delay since the procedure.11, 40 For primary prevention there are no studies showing that interruption of aspirin carries an increased perioperative risk except in diabetic patients.

Dual antiplatelet therapy is recommended for 2 weeks after simple angioplasty, 4 to 6 weeks after BMS and at least 12 months after DES (see table).10, 40 All elective surgical procedures should be postponed beyond these delays. Only vital surgery should be performed when patients are still on aspirin and clopidogrel.11 Unless the hemorrhagic risk is excessive, dual therapy should not be interrupted for surgery during the first 6 weeks after BMS and at least 12 months after DES. For secondary prevention of stroke neither clopidogrel or dual therapy has been proven more efficacious than aspirin alone or aspirin and dipyridamole. Thus, it seems safe to stop clopidogrel but not aspirin 5 days before surgery.9, 18

	Recommended duration of dual antiplatelet therapy after PCI
	Aspirin (75–325 mg/day)	Lifelong without interruption
	Clopidogrel (75 mg/day):
	Angioplasty without stenting	2–4 Wks
	PCI + bare metal stent	4–6 Wks
	Myocardial infarction	3–6 Mos
	ACS (unstable)	12 Mos
	PCI + DES	12 Mos (greater than 12 mos for high risk stents)
	Recommended delays for noncardiac surgery after coronary revascularization
	Dilatation without stenting	2–4 Wks (only vital surgery)
	Bare metal stent	6 Wks vital surgery, 3 mos elective surgery
	CABG	6 Wks vital surgery, 3 mos elective surgery
	Drug-eluting stent	12 Mos elective surgery

Modified from Chassot et al.40, 46

If clopidogrel must be interrupted 5 days preoperatively for surgical reasons in high cardiac risk situations, aspirin should be continued without interruption.43 Clopidogrel could be substituted with a short-acting equivalent drug. Although not proven by a controlled trial, a bridge with an anti-GP IIb/IIIa agent such as eptifibatide or tirofiban (half-life 2 hours) has been described as a possible substitution for clopidogrel.44, 45 The substance is administered in continuous infusion for 3 to 5 days before the operation and stopped 6 hours preoperatively. There is no scientific evidence to support the use of heparin in preventing intraoperative stent thrombosis.10, 11 Preoperative low molecular weight heparin has been shown to have no effect because antithrombins have no antiplatelet activity.7 A few authors have designed working algorithms based on expert opinions but these are somewhat arbitrary. The algorithm we propose is adapted with minor modifications from previous publications (fig. 3).40, 46

View Large Image Download to PowerPoint

Figure 3. Algorithm for preoperative treatment of patients on AP therapy adapted and updated from Chassot et al.40 1 Aspirin given after previous event (MI, ACS, PCI and stent, CABG, peripheral arterial disease and stroke) or for documented cardiovascular disease, and high risk patients include diabetics and those with multiple risk factors. 2 Multiple stents, long stents, proximal location (left main) and bifurcation lesions, and patients with previous stent thrombosis. 3 Operations mandatory for long-term survival. Recommended delay between drug interruption and surgery is 5 days for aspirin and clopidogrel. 4 Risk-to-benefit ratio of maintaining vs withdrawing aspirin, clopidogrel and prasugrel must be evaluated for each case, and withdrawing postoperative reinstitution (with clopidogrel loading dose 300 mg) within 24 hours is important. 5 Stop clopidogrel for 5 days and substitute with 3 to 5 days of intravenous perfusion of eptifibatide or tirofiban. SCA, sudden cardiac arrest.

Since there is no antidote to antiplatelet agents, termination of the inhibitory effect of aspirin and clopidogrel relies on the renewal of platelets (10% a day). It is usually estimated that the plasma level of a substance is negligible after 3 half-lives. Therefore, 24 hours after the last intake of clopidogrel (half-life of active metabolite 8 hours) and 6 hours after the perfusion of eptifibatide or tirofiban (half-life 2 hours) there is no residual antiplatelet activity in the plasma. Therefore, during surgery the platelets transfused after these delays will function normally. In this situation the risk is to trigger thrombus formation inside the stents because of platelet over transfusion. To prevent this potentially catastrophic event it is necessary to accept some degree of platelet dysfunction.

Clopidogrel therapy needs to be restarted within the first 24 hours after the operation because of concerns about stent thrombosis during the postoperative phase of hypercoagulability.11 It might be safer to reinitiate clopidogrel with a 300 mg loading dose which reduces the time to achieve maximal platelet inhibition to 4 hours and decreases the risk of hyporesponsiveness linked to the competition of other drugs with the hepatic cytochromes oxidizing clopidogrel into its active metabolite. Postoperative stent thrombosis, usually manifested as an acute MI and cardiogenic shock, is best treated with immediate PCI and dilation. However, the survival rate with this strategy is only 65%.47 The use of stent is questionable because patients are at the peak of the acute inflammatory reaction and platelet hyperactivity.

Strategies to Prevent Stent Thrombosis 

In most cases urological surgery can be postponed by weeks or even months after the high risk period. This strategy alone will move most patients to the low risk group requiring continued single agent antiplatelet therapy (aspirin or clopidogel). Furthermore, the operation can possibly be deferred by active surveillance strategies instead of surgery or by closely following small renal masses by computerized tomography instead of acute intervention. These options may be valuable in patients with recent cardiovascular events. Nevertheless, many patients will still require continued antiplatelet therapy during surgery. Since clear guidelines have not yet been established, the surgeon must base the decision on sparse reports.2, 45

Another possible strategy to prevent postoperative stent thrombosis is to avoid preoperative revascularization when possible because it does not benefit patients with stable48 or even severe49 coronary artery disease compared to an adequate medical therapy with tight heart rate control. In case of urgent or semi-urgent surgery the risk of operating under maximal medical protection (β-blocker, aspirin, clopidogrel, statin) is less than that of operating in the early phase after coronary revascularization.

Some patients may require coronary revascularization because they have an unstable coronary syndrome or severe ischemia on stress testing and present with a surgical pathology necessitating a mandatory operation. The strategy consists of adapting the type of revascularization to the possible delays for noncardiac surgery (fig. 4). If the operation can be delayed for 6 to 8 weeks, the best option is a PCI with BMS or surgical revascularization (CABG). If an operation is required within 2 to 4 weeks, the only possibility is to perform a balloon angioplasty without stenting or refrain from any revascularization.11, 40 In case of emergency surgery must be performed using pharmacological cardioprotection except when vascular surgery can be combined with simultaneous CABG. DES implantation is out of the question because the optimal delay after DES placement is more than 1 year.

View Large Image Download to PowerPoint

Figure 4. Algorithm of strategies to decrease risk of stent thrombosis in case of coronary revascularization and urgent or semi-urgent noncardiac surgery. Asterisk indicates except in closed space surgery (intracranial neurosurgery, surgery of medullary canal, surgery of posterior chamber of eye), and invasive surgery with severe bleeding and difficult hemostasis.

Another strategy is to maintain dual antiplatelet therapy in the preoperative period and resume the treatment as soon as possible after surgery. This strategy applies to high cardiac risk situations (less than 6 weeks after ACS, MI, BMS or stroke, or less than 12 months after DES placement). In intermediate cardiac risk situations (6 to 24 weeks after ACS, MI or BMS, or greater than 12 months after high risk DES) it is advisable to maintain dual therapy if the risk of surgical bleeding is low but it is possible to stop clopidogrel and maintain only aspirin for operations with high hemorrhagic risk.43

Future Trends 

Of the new agents prasugrel is a thienopyridine which acts like clopidogrel but is more potent and presents a much lower rate of nonresponse. It is 2-fold more efficacious than clopidogrel for preventing thrombosis after PCI and stenting but it increases the nonsurgical bleeding rate by 32% compared to clopidogrel (2.4% vs 1.8%), which might raise problems in the surgical setting.50 All data about DESs concern first generation stents coated with sirolimus or paclitaxel. New devices have been introduced recently in cardiology, including everolimus-eluting, zotarolimus-eluting and bioabsorbable stents. However, data on long-term outcome of these stents are insufficient to justify a modification of the recommendations published for first generation DESs.

Conclusions 

Recent data on perioperative use of antiplatelet agents are a warning signal against their withdrawal before surgery when prescribed for secondary prevention after ACS, MI, coronary stents or stroke. In general the risk of coronary thrombosis largely outweighs the risk of surgical hemorrhage. However, this issue has not yet been clearly evaluated in urological patients and in many instances during urological surgery the risk of bleeding might be exceedingly dangerous. In this problematic situation a thorough dialogue among surgeon, cardiologist and anesthesiologist is essential to determine all risk factors, and define the best possible strategy for each patient. Although ethical difficulties prevent placebo controlled, randomized trials, further prospective clinical studies are needed to define the urological surgeries that can be performed safely during continued antiplatelet therapy.