Microsoft word - periop anticoag review _w97_.doc

Perioperative Anticoagulation Management in Patients who are
Receiving Oral Anticoagulant Therapy:
A Practical Guide for Clinicians
From the Department of Medicine, McMaster University and St. Joseph  s Healthcare, Hamilton, Canada.
anticoagulation, surgery, mechanical heart valve, atrial fibrillation 50 Charlton Ave. East, Hamilton, Ontario, Canada, L8N 4A6 Abstract
The management of patients who require temporary interruption of oral anticoagulant therapy because of surgery or other invasive procedures is a clinically important topic because ofthe increasing prevalence of patients who are receiving oral anticoagulants, and the availability oflow-molecular-weight heparins, which allow out-of-hospital perioperative anticoagulation. Theoptimal management of such patients has been hampered by the lack of well-designed prospectivestudies investigating the efficacy and safety of different perioperative management strategies. Thetwo main issues that need to be considered in perioperative anticoagulant management is thepatient  s risk of thromboembolic event when anticoagulant therapy is interrupted, and the risk ofbleeding that is associated with the surgery or procedure. An assessment of these factors willdetermine the perioperative management approach. The objectives of this review is to focus onpractical issues relating to perioperative anticoagulation; and the implementation of aperioperative anticoagulation management approach that can be used in everyday clinical practice.
Case Presentation
A 60 year-old woman, weighing 70 kg, with rheumatic valvular heart disease and a mechanical bileaflet mitral valve is scheduled to undergo elective sub-total colectomy foradenocarcinoma of the sigmoid colon. Her past medical history includes chronic atrial fibrillation,hypertension, and steroid-dependent chronic obstructive pulmonary disease. Her medications areaspirin, warfarin, diltiazem, ramipril, prednisone, and inhaled bronchodilators. There is noprevious history of stroke, other cardiovascular events, or renal insufficiency.
Introduction
The management of patients, as described above, who require temporary interruption of oral anticoagulant therapy because of surgery or an invasive procedure is a frequentlyencountered but under-investigated clinical problem. In North America, there are over one millionpeople with a mechanical prosthetic heart valve, chronic atrial fibrillation, or venousthromboembolism who are receiving long-term warfarin therapy (1). With an aging population,the number of warfarin-users will increase (2), and since elderly people over 65 years of age aretwo times more likely to require surgery than non-elderly people (3), the clinical scope ofperioperative anticoagulant management will increase considerably.
Surprisingly, few studies have investigated the efficacy and safety of different perioperative anticoagulant management strategies (4-12). The traditional approach is tohospitalize patients for 3 to 4 days before and 3 to 4 days after surgery to administer intravenousunfractionated heparin (UFH) during the period that warfarin therapy is temporarily interruptedand the international normalized ratio (INR) is subtherapeutic (13-15). The emergence of low-molecular-weight heparins (LMWHs), which can be administered in a fixed, weight-adjusted,subcutaneous dose and without laboratory monitoring (16), obviates the need for in-hospitalperioperative anticoagulation. This approach is appealing because of its convenience, and thepotential to substantially reduce health care costs (17).
The rationale for administering full-dose (or therapeutic-dose) LWMH or UFH as bridging anticoagulant therapy while warfarin is temporarily interrupted is to shorten the time period thatpatients are not receiving therapeutic anticoagulation and, therefore, to minimize the risk ofthromboembolic events (13-15). In addition, after warfarin or antiplatelet therapy is stopped, atransient hypercoagulable state may develop due to a rebound increase in thrombin generation(18,19), or platelet activation (20). Despite these considerations, the efficacy, in terms ofpreventing perioperative thromboembolism, and safety, in terms of minimizing postoperativebleeding, of bridging anticoagulant therapy has been neither substantiated nor challenged by well-designed clinical trials. Authorities that include the American College of Cardiology /AmericanHeart Association Task Force on Practice Guidelines and the American College of ChestPhysicians recommend that bridging anticoagulant therapy with UFH or LMWH should beconsidered for most patients who require temporary interruption of warfarin therapy (21,22).
However, the need for bridging anticoagulant therapy is not universally accepted. Some authorities have argued that, with the exception of a minority of high-risk patients who have had athromboembolic event within the preceding month, the risk of thromboembolism during theperiod of warfarin interruption has been overstated, and is outweighed by the risk ofpostoperative bleeding due to bridging anticoagulation (23,24). Other authorities do not specifywhich patients should receive bridging anticoagulant therapy (25).
The lack of consistent recommendations regarding perioperative anticoagulant management is reflected in the variability of physicians  practice patterns. In a recent survey ofperioperative anticoagulation practices of 473 Canadian internists (26,27), full-dose intravenousUFH was preferred by 60% of respondents, full-dose LWMH was preferred by 15% ofrespondents, and several other anticoagulation strategies were preferred by the remainingrespondents. It is noteworthy that the physicians  management preferences were driven more bythe need to prevent perioperative thromboembolism than concerns about postoperative bleeding.
In the absence of evidence-based, graded, practice recommendations, several recent reviews have provided general guidelines for perioperative anticoagulation (28-32). However,perhaps what is lacking in these reviews is a practical management approach, with specific patientcare guidelines that can be used by clinicians in everyday clinical practice. Therefore, theobjectives of this review are: 1) to address practical issues relating to perioperativeanticoagulation; and 2) to provide a perioperative anticoagulation management approach that canbe integrated into everyday clinical practice.
Assessment of Perioperative Thromboembolic Risk
The assessment of a patient  s baseline thromboembolic risk is based on three factors: 1) the clinical indication for anticoagulation; 2) the presence of additional thromboembolic riskfactors; and 3) the clinical consequences of a thromboembolic event. This review will focus onpatient groups who are frequently assessed for perioperative anticoagulation management.
Patients with a Mechanical Prosthetic Heart Valve Anticoagulant therapy is required in patients with a mechanical prosthetic heart valve to prevent stroke and systemic embolism, and to prevent valve thrombosis, which is associated witha 15% mortality rate (33,34). Patients with a prosthetic mitral valve, a caged-ball valve, and twoprosthetic heart valves are at highest risk for thromboembolic events (35). Patients with aprosthetic aortic valve and two or more thromboembolic risk factors may be considered atmoderate risk for thromboembolic events, whereas patients with a prosthetic aortic valve and lessthan two thromboembolic risk factors are at lowest risk. The most thrombogenic prosthetic heartvalve is the caged-ball type (eg., Starr-Edwards), followed by the tilting disc type (eg., Bjork-Shiley, Lillehei-Kaster), and bileaflet type (eg., St. Jude, Carbomedics), which is the leastthrombogenic. Prosthetic mitral valves are more thrombogenic than aortic prostheses because ofgreater vascular stasis around the mitral valve (35).
Providing reliable estimates of the absolute risk of thromboembolic events in such patients during temporary interruption of warfarin therapy is problematic because most data are based on
studies performed 20 to 30 years ago, and involve patients with mainly first-generation prosthetic
heart valves. Nonetheless, it is estimated that the incidence of a thromboembolic event in patients
with a prosthetic heart valve who are not receiving anticoagulant therapy is between 9 and 22%
per year (36,37). If one interpolates this risk to a 6 to 8 day perioperative period, when warfarin
therapy is interrupted, this corresponds to an absolute risk of thromboembolism of 0.17 to 0.42%.
Few studies have investigated the perioperative risk of thromboembolism so as to corroborate
these risk estimates (Table 1). In four studies involving 271 patients with mostly first-generation
prosthetic heart valves who underwent 300 procedures, 4 (1.3%) patients had a perioperative
thromboembolic event despite bridging anticoagulant therapy (7-10). In a retrospective study of
235 patients with newer prosthetic valve types (191 bileaflet, 51 tilting disc), of whom 164 had
bridging anticoagulant therapy with full-dose UFH or LMWH, a higher rate of thromboembolism
was reported (11). In this study, 4 of 177 (2.3%) patients with a aortic valve prosthesis, 9 of 51
(18%) patients with a mitral valve prosthesis, and 3 of 7 (43%) patients with two prosthetic
valves had a thromboembolic event. However, limitations of this study included inadequate
information about clinical follow-up and the adequacy of perioperative anticoagulation. Two
small, but well-designed, prospective cohort studies of bridging anticoagulant therapy with full-
dose LMWH reported one thromboembolic event in 24 (4.2%) patients (5,6). Overall, the
discrepant findings from these studies underscore the need for large, well-designed, prospective
studies to provide reliable estimates of the risk of thromboembolism in patients with a mechanical
heart valve who have interruption of warfarin and receive bridging anticoagulant therapy. Based
on the available evidence, a classification scheme that stratifies patients with a mechanical heart
valve according to the thromboembolic risk, and a suggested anticoagulation management
strategy for each risk category, is provided in Table 2.
In patients with chronic atrial fibrillation, criteria for thromboembolic risk classification are well-established (38,39). Patients with a previous stroke or transient ischemic attack are at highestrisk of recurrent stroke, with an incidence of 12 to 15% per year (40,41). Patients with lone atrialfibrillation, who are 65 years old or younger and have no additional stroke risk factors have the lowest risk of stroke, with an incidence of <1% per year (42). However, such patients comprise
less than 2% of all patients with chronic atrial fibrillation (42). Most patients with chronic atrial
fibrillation fall between these two risk extremes, and have an incidence of stroke of 3 to 7% per
year (41). In this moderate risk group, the presence of additional risk factors can be used to
further stratify patients according to thromboembolic risk. There are four risk factors, in addition
to a previous stroke that are associated with an increased risk of stroke (38,39): i) age >75 years;
ii) hypertension; iii) diabetes mellitus; and iv) left ventricular dysfunction. These factors are
similarly weighted in terms of their stroke risk (Table 3).
During temporary interruption of warfarin therapy, the risk of stroke and other thromboembolic events can be interpolated from the aforementioned data, in high-, moderate-,
and low-risk patients wih chronic atrial fibrillation. In high-risk patients, the absolute risk of
thromboembolism is between 0.28 and 0.38%, in moderate risk patients it is 0.06, and 0.15%, and
in low-risk patients it is between 0.02 and 0.04%. However, only two studies, involving 20
patients, have investigated the perioperative clinical course of such patients (5,6). Consequently,
there is inadequate data to provide reliable risk estimates for perioperative thromboembolism in
patients with chronic atrial fibrillation. In Table 4, we provide a classification scheme that
stratifies patients with a chronic atrial fibrillation according to the thromboembolic risk, and a
suggested anticoagulation management strategy for each risk category.
In patients with venous thromboembolism, there are no data in regard to the risk of recurrent disease if warfarin therapy is temporarily interrupted. Based on studies investigating the
clinical course of treated venous thromboembolism, the risk of recurrent disease is likely highest if
warfarin therapy is interrupted soon after the start of treatment (43,44). In one study of 1,021
patients with venous thromboembolism, 42 of 58 (72%) recurrent events occurred during the first
three weeks after diagnosis, and the incidence of recurrent disease decreased sharply thereafter
(43). The risk of recurrence is higher in patients with cancer, chronic diseases, and
antiphospholipid antibodies (39-42). In patients who have completed 6 to 12 weeks of
anticoagulant therapy, the risk of recurrence depends mainly on the disease etiology. Idiopathic
(or unprovoked) venous thromboembolism is associated with a recurrence risk of 10 to 27% per
year (46,47), whereas venous thromboembolism occurring after exposure to a reversible risk
factor, such as surgery, is associated with a 2 to 5% per year risk of disease recurrence (47-49). It
is probable that patients with previous postoperative venous thromboembolism will be at
increased risk of disease recurrence after surgery, although there are no data to provide reliable
estimates of this risk. Based on these considerations, we provide a classification scheme in Table
5 that stratifies patients with venous thromboembolism according to the risk of disease
recurrence, and a suggested anticoagulation management strategy for each risk category.
Assessment of Postoperative Bleeding Risk
The assessment of postoperative bleeding risk is based the following considerations: 1) the adequacy of postoperative hemostasis; and 2) the risk of bleeding and clinical consequences ofbleeding associated with surgery or an invasive procedure.
Determining the adequacy of postoperative hemostasis during after surgery is a subjective assessment that is based on the amount of bleeding at the surgical site, ongoing blood loss into asurgical drain, and the presence of clinical features suggestive of bleeding. Excessive wound-related bleeding is characterized by repeatedly blood-soaked wound bandages, and ongoing bloodloss of >250 ml/12 hours into a surgical wound drain (50,51). In general, patients who were notreceiving antithrombotic agents and had normal hemostatic function before surgery should haveadequate hemostasis 12 to 24 hours after surgery. Risk of Bleeding Associated with Surgery or an Invasive Procedure In patients who undergo intraabdominal, intrathoracic or orthopedic surgery, there should be adequate postoperative hemostasis by 24 hours after surgical closure. However, there are
several surgical procedures that can be associated with significant postoperative bleeding, despite
adequate surgical technique. A bleeding risk classification scheme for various procedures is
provided in Table 6. Urologic procedures, such as prostatectomy, bladder tumour ablation and
renal biopsy, are associated with considerable postoperative bleeding, partly due to the presence
of urokinase that is produced by genitourinary epithelium (52,53). Patients who undergo
percutaneous renal biopsy are a particular concern because retroperitoneal bleeding may be
clinically silent until there is major blood loss (54). Gastrointestinal procedures that are associated
with an increased risk of bleeding include bowel resection and anastomosis, particularly in patients
who are receiving corticosteroids, which may delay anastomotic wound healing (55,56).
Endoscopic removal of a colonic polyp also may be associated with considerable bleeding (31,57).
Coronary artery bypass surgery is associated with an increased risk of mediastinal bleeding
because of coagulation factor dilution and heparinization that occurs during cardiopulmonary
bypass (58,59). Neurosurgical procedures that involve the brain or spinal cord, in general, are
associated with an increased risk of potentially serious intracranial or epidural bleeding (60).
Perioperative Management with Bridging Anticoagulant Therapy
This section will focus on how to implement bridging anticoagulant therapy with out-of- hospital full-dose LMWH and will refer to a suggested perioperative patient care path outlined in
the Figure. The use of in-hospital intravenous UFH as an alternative to LMWH also will be
discussed. In general, bridging anticoagulant therapy should be considered in patients at high-risk
for thromboembolic events and, possibly, for moderate-risk patients. In patients at low-risk for
thromboembolic events, in whom bridging anticoagulant therapy is optional, the patient care path
can be used but with omission of the section on full-dose LMWH or UFH administration.
i) Stopping antithrombotic drugs. There is a increased risk of intra- and post-operative bleedingwhen surgery is performed in patients who are receiving an anticoagulant (61,62), or antiplateletdrug (63-65). To ensure normal hemostatic function at the time of surgery, antiplatelet drugsshould be stopped 7 to 10 days before surgery, as aspirin, ticlopidine and clopidogrel allirreversibly inhibit platelet function for the 7 to 10 day lifespan of a platelet (66). In patients whoare receiving warfarin therapy with a target INR of 2.0 to 3.0, stopping warfarin 5 days beforesurgery will, in the vast majority of cases, ensure a normal INR at the time of surgery (67).
Elderly patients appear to require a longer time for the INR to normalize after warfarin is stopped(68), and in patients with a target INR of 2.5 to 3.5, it is reasonable to stop warfarin 6 days beforesurgery.
ii) INR measurements before surgery. In addition to INR testing on the day warfarin therapy isstopped, INR testing should be performed on the day before surgery to ensure that it is normal(INR <1.3) or near-normal (INR = 1.4). Patients who undergo surgery with an INR of >1.5 are atincreased risk of postoperative bleeding complications (62). If the INR is >1.5 on the day beforesurgery, administering low-dose oral vitamin K (i.e., 1 to 2 mg) will ensure a normal INR at thetime of surgery, and is unlikely to cause resistance to re-anticoagulation when warfarin is resumedafter surgery (69,70).
iii) Preoperative bridging anticoagulant therapy.
Bridging anticoagulant therapy with LWMH is started, typically, 3 to 4 days before surgery, when a patient  s INR is below or is expected to be below the lower limit of thetherapeutic range. In patients who are receiving warfarin with a target INR of 2.5 to 3.5, bridginganticoagulant therapy is started when the INR is <2.5, and in patients with a target INR of 2.0 to3.0, it is started when the INR is <2.0. In many instances, however, the caregiver will not knowwhen the INR drifts below the therapeutic range unless daily INR testing is performed, which isimpractical. Thus, it is reasonable to empirically start LMWH therapy two days after warfarin isstopped. If once-daily LMWH is used as bridging anticoagulant therapy (eg., tinzaparin, 175IU/kg once-daily), it should be administered in the mornings, and with last preoperative dosegiven on the day before surgery. If twice-daily LMWH is used (eg., enoxaparin, 1 mg /kg twice-daily), the evening dose before surgery should be omitted. With either dosing regimen, the lastdose of LMWH should be administered at 20 to 24 hours before surgery, to eliminate thelikelihood of a residual anticoagulant effect at the time of surgery (16).
i) Resumption of LMWH. The decision to resume LMWH after surgery is based on whether thereis adequate postoperative hemostasis and the bleeding risk associated with surgery. If there isongoing bleeding, detected by accumulation of blood into a surgical drain, the resumption ofLMWH should be deferred until the bleeding has subsided. Most postoperative bleeding that isdue to delayed wound healing will resolve within 24 hours of surgery. If there is adequatepostoperative hemostasis, the decision to resume anticoagulant therapy is based on the bleeding risk associated with the surgical procedure. In patients undergoing surgery that is associated with
a high risk of bleeding (Table 6), the resumption of LMWH should be deferred for at least 24 to
48 hours after surgery and, preferably, after consultation with the surgeon. The initial treatment of
LMWH should be low-dose, which is used as prophylaxis against deep venous thrombosis (Table
7). This low-dose regimen can be started on the first or second day after surgery. The use of full-
dose (therapeutic-dose) LMWH should be avoided for 48 to 72 hours after surgery, or may not
be used at all if there is ongoing concern about postoperative bleeding. In patients undergoing
surgery that is associated with a moderate or low risk of bleeding, low-dose LMWH can be
resumed on the evening after surgery. If this treatment is well-tolerated, and without bleeding
complications, subsequent doses of LMWH can be administered with a full dose regimen, starting
24 to 48 hours after surgery (Table 8).
ii) Resumption of warfarin. In most patients, warfarin can be restarted the evening after surgery.
A minimal anticoagulant effect of warfarin will not occur for at least 24 hours after the initial doseof warfarin, and a therapeutic anticoagulant effect will not occur for 4 to 5 days after the start ofwarfarin therapy (67). The dose of warfarin can correspond to the dose the patient usuallyreceives on that day of the week. If a patient received high-dose vitamin K (i.e., 5 to 10 mg)before surgery, this may result in resistance to re-anticoagulation when warfarin is resumed.
Because it is difficult to predict the warfarin dose requirements of patients who have received ahigh dose of vitamin K, it is reasonable to double the patient  s warfarin dose for two consecutivedays after surgery. In addition, INR monitoring can be performed more frequently after surgery todetect persistently low INR levels and to adjust the warfarin dose accordingly. If low-dosevitamin K (i.e., 1 to 2 mg) is administered before surgery, resistance to re-anticoagulation isunlikely (69,70). Nonetheless, it is reasonable to double the first dose of warfarin in such patients,and to resume the patient  s usual dose of warfarin on the following day.
iii) Stopping LMWH. LMWH should be stopped when the INR is within the target therapeuticrange. In most patients with a target INR of 2.0 to 3.0, LMWH or UFH will be required for 3 to4 days after surgery, and in most patients with a mechanical heart valve and a target INR of 2.5 to3.5, LMWH or UFH will be required for 4 to 5 days after surgery.
iv) Patients with unexpected postoperative bleeding. Major bleeding, defined as bleedingassociated with transfusion of >2 units of packed red blood cells, re-operation, or bleeding intothe intracranial, intrathoracic or retroperitoneal cavity (71), can be separated into surgical andnon-surgical bleeding. Surgical bleeding at the wound site is the most common source of bleeding(72-74), and usually results from delayed wound healing or intraoperative blood vessel injury.
Bleeding also can occur from stress-induced peptic ulceration or a hemorrhagic stroke. Theanticoagulant management in such patients depends on the cause and location of bleedingalthough, in general, anticoagulants should be withheld until the bleeding source has beenidentified and treated. If the cause of bleeding is reversible, as with the repair of a severed bloodvessel that inadvertently occurred during surgery, anticoagulant therapy can be resumed probablywithin 24 hours. In patients with gastrointestinal bleeding, anticoagulants can be withheld for aslittle as two days in patients with a self-limiting Mallory-Weiss tear, to as long as 21 days inpatients with a large gastric ulcer (75,76). In patients with intracranial bleeding, resumption ofanticoagulants can occur within one to four weeks, depending on the extent of bleeding and radiologic evidence of healing (77-79).
Unfractionated Heparin as Bridging Anticoagulant Therapy
If UFH is used for bridging anticoagulant therapy, the conventional approach is to hospitalize patients for 3 to 4 days before and 3 to 4 days after surgery to administer intravenousUFH, with dose adjustments to achieve a targeted activated partial thromboplastin time (aPTT) of1.5- to 2-times the control aPTT (80). The preoperative infusion of UFH should be stopped atleast four hours before surgery, and resumed 12 to 24 hours after surgery when there is adequatepostoperative hemostasis. To minimize the risk of postoperative bleeding, it is reasonable to avoidthe use of a standardized heparin nomogram because of the unpredictable dose-response of UFH,and the potential that patients may have persistently high aPTT levels (i.e., >150 seconds).
Instead, one might consider a more conservative dosing regimen of intravenous UFH, with atarget aPTT of 45 to 60 seconds rather than the target aPTT of 60 to 80 seconds that used withthe standardized heparin nomogram (80). Although this approach has not been tested inprospective clinical trials, there is evidence that a persistently increased aPTT in heparin-treatedpatients is associated with an increased risk of bleeding (81,82). A drawback of this approach isthe need for periodic adjustment of the UFH infusion rate. An alternative management approach isout-of-hospital, twice-daily subcutaneous injections of UFH (83). However, this approachrequires daily, mid-interval, aPTT testing to be performed 6 hours after the morning dose of UFHto monitor the anticoagulant effect and make dose adjustments as required.
Management of Specialized Patient Groups
In patients with significant renal insufficiency, defined by a serum creatinine >150 mmol/L or a creatinine clearance <40 ml/min, UFH is the anticoagulant of choice because it is not clearedprimarily by the kidney (16). LMWH should be avoided in patients with renal insufficiencybecause it is cleared primarily by the kidney. Furthermore, if LMWH is used, its bioaccumulationof LMWH may be undetected, as the aPTT will not be increased, thereby increasing the risk ofintra- and post-operative bleeding (16,84). If LMWH is used in patients with renal insufficiency,the anticoagulant effect of LMWH should be measured with an anti-factor Xa level done 4 hoursafter the LMWH dose, with a targeted therapeutic anti-factor Xa level of 0.5 to 1.0 units/ml (16).
Patients with Previous Heparin-Induced Thrombocytopenia Heparin-induced thrombocytopenia (HIT) is a rare but serious complication of UFH therapy, and to a lesser extent LMWH therapy, that is associated with venous limb gangrene,stroke and myocardial infarction (85). UFH and LMWH should be avoided in patients with HIT.
Danaparoid, a low-molecular-weight heparinoid with minimal cross-reactivity with UFH, can besafely administered in patients with HIT. The initial low dose of danaparoid is 750 IU,administered subcutaneously on the evening after surgery, with subsequent doses increased to1,250 IU, twice-daily. Alternatively, lepirudin, a direct thrombin inhibitor, can be administered topatients with HIT. Lepirudin is given initially as a loading dose, 0.4 mg/kg intravenously over 15 to 20 minutes, followed by a continuous infusion at 0.15 mg/kg/hr for 2 to 10 days, and can alsobe administered subcutaneously (86).
Patients with Spinal Anesthesia or Continuous Epidural Analgesia The anticoagulant management in patients who had a spinal puncture before surgery is problematic because of the risk of a spinal epidural hematoma, a rare but devastating complication(87,88). In patients who have spinal anesthesia, with epidural catheter removal immediately aftersurgery, it is safe to resume anticoagulant therapy within 12 hours after surgery. If the epiduralcatheter placement was traumatic, the resumption of anticoagulants should be delayed for at least24 hours after surgery. In patients who have an indwelling epidural catheter after surgery toadminister analgesia, anticoagulants, in general, should be withheld until the epidural catheter isremoved. However, co-administration of LMWH and continuous epidural analgesia may beconsidered if the following criteria are satisfied (89): i) epidural catheter placement was non-traumatic; ii) low-dose, once-daily, LMWH is administered; iii) warfarin is started after theepidural catheter is removed; iv) the epidural catheter is removed during the trough anticoagulanteffect of LMWH, corresponding to 18 to 22 hours after the preceding dose; v) other drugs thateffect haemostasis, such as aspirin or NSAIDs, are avoided until the epidural catheter is removed.
Case Discussion
This case represents a patient who is considered at high risk for perioperative thromboembolic events because of a mitral valve prosthesis, and two additional risk factors(chronic atrial fibrillation, hypertension). The surgery is associated with a moderate bleeding risk.
However, corticosteroid use may impair wound healing, particularly at the colostomy site, withthe potential to increase wound-related bleeding. The overall management approach consists ofbridging anticoagulant therapy with out-of-hospital , full-dose, LMWH, and cautious resumptionof anticoagulants after surgery.
Preoperative management. The patient is instructed to stop aspirin 10 days before surgery, and tohave INR testing 6 days before surgery. One that day, the INR level is 3.9, and the patient isinstructed to stop warfarin. Rather than repeating the INR daily to determine when the INR driftsbelow the therapeutic range (i.e., INR <2.5), out-of-hospital LMWH is started empirically 3 daysbefore surgery, with dalteparin 100 IU twice-daily, subcutaneously. INR testing is repeated theday before surgery, and is elevated at 1.6. To ensure the INR has normalized by the time ofsurgery, the patient receives vitamin K, 1 mg, orally. The last preoperative dose of dalteparin isadministered on the morning of the day before surgery to ensure no residual anticoagulant effectat the time of surgery. The INR is re-checked and is 1.3 on the morning before surgery.
Intraoperative course. Intraoperative spinal epidural anesthesia is administered, and the surgery isuneventful. Because of concerns with parenteral analgesia use in a patient with chronicobstructive lung disease, the epidural catheter remains is place to administer epidural analgesia.
Postoperative management. In the immediate postoperative period, there appears to be adequatehemostasis although there is ongoing drainage of serosanguinous fluid from an abdominal drain, with 250 ml having accumulated within 6 hours after surgical closure. The indwelling epiduralcatheter is to be removed the day after surgery. Warfarin is resumed on the evening of the day ofsurgery, with sips of water, starting with the patient  s usual dose for that day of the week. On the1st postoperative day, low-dose LMWH is administered (i.e., dalteparin 5,000 IU once-daily),three hours after removal of the epidural catheter. On the 2nd postoperative day, there is noongoing blood loss from the intraabdominal drain, and LMWH is increased to full-dose treatment(i.e., dalteparin 7,000 IU twice-daily). Aspirin is also restarted on this day. On the 5thpostoperative day, when the INR is 2.5, full-dose LMWH is stopped after the evening dose.
References
Madura JA, Rookstool M, Wease G. Themanagement of patients on chronic coumadintherapy undergoing subsequent surgical Top 200 brand-name drugs by prescription in procedures. Am Surgeon 1994;60:542-547.
1999. Drug Topics 2000: 69-70. Available at: anticoagulation in patients with prosthetic heart Furberg CD, Psaty BM, Manolio TA, Gardin JM, valves undergoing non-cardiac operations.
Smith VE, Rautaharju PM. Prevalence of atrial Bryan AJ, Butchart EG. Prosthetic heart valves Cardiovascular Health Study). Am J Cardiol cardiac surgery. Br J Surg 1995;82:577-578.
Politser P, Schniederman D. American College of 16. Hirsh J, Warkentin TE, Shaughnessy SG, et al.
Surgeons: Socioeconomic Factbook for Surgery Heparin and low-molecular-weight heparin: 1990. Chicago, American College of Surgeons, mechanism of action, pharmacokinetics, dosing, discontinuation of oral anticoagulants: role of low Goldstein JL, Larson LR, Yamashita BD, Fain molecular weight heparin (dalteparin).
Circulation 2000;102 (suppl II):826 (abstract).
heparin versus unfractionated heparin in the Spandorfer JM, Lynch S, Weitz HH, Fertel S, colonoscopy peri-procedure period: a cost Merli GJ. Use of enoxaparin for the chronically procedures. Am J Cardiol 1999;84:478-480.
Genewein U, Haeberli A, Straub PW, et al.
Rebound after cessation of oral anticoagulant therapy: the biochemical evidence. Br J Haem periprocedure anticoagulation for patients on warfarin and at high risk of thrombosis. Ann Hypercoagulable state and thromboembolism Katholi RE, Nolan SP, McGuire LB. Subsequent following warfarin withdrawal in post-myocardial infarction patients. Eur Heart J 1991;12:1225- thromboembolism or hemorrhage. Am Heart J Valles J, Aznar J, Santos T, et al. Platelet function in patients with chronic coronary heart disease on long-term anticoagulant therapy: effect of management of anticoagulation during noncardiac anticoagulant therapy. Haemostasis 1993;23:212- operations in patients with prosthetic heart Bonow RO, Carabello B, De Leon AC, Edmunds LH Jr, Fedderly BJ, Freed MD, et al. ACC/AHA guidelines fro the management of patients with anticoagulation therapy is surgical patients with valvular heart disease. A report of the American cardiac valve prostheses. JAMA 1978;239:738- Association Task Force on Practice Guidelines Darmon D, Enriquez-Sarano M, Acar J. Cardiac (Committee on Management of Patients with complications after subsequent non-cardiac Valvular Heart Disease). JACC 1998;32:1558- operations in patients with non-biological prosthetic heart valves. Eur J Cardiol 1983;4 Ansell J, Hirsh J, Dalen J, Bussey H, et al.
Managing oral anticoagulant therapy. Chest Carrel TP, Klingenmann W, Mohacsi PJ, et al.
Perioperative bleeding and thromboembolic risk during non-cardiac surgery in patients with anticoagulation before and after elective surgery.
Hewitt RL, Chun KL, Flint LM. Current clinical concepts in perioperative anticoagulation. Am Manley HJ, Smith JA, Garris RE. Subcutaneous enoxaparin for outpatient anticoagulation therapy Gohlke-Bäwolf C, Krozingen B, et al. Guidelines in a patient with an aortic valve replacement.
for prevention of thromboembolic events in anticoagulation in patients with a mechanical Douketis JD, Crowther MA, Cherian S, Kearon heart valve who are undergoing elective non- C. Physician preferences for perioperative cardiac surgery. Chest 1999;116:1240-6.
Douketis JD, Crowther MA, Cherian SS.
Investigators. Stroke 1999;30:1223-1229.
Perioperative anticoagulation in patients with Secondary prevention in non-rheumatic atrial chronic atrial fibrillation undergoing elective fibrillation after transient ischemic attack or surgery: Results of a physician survey. Can J minor stroke: EAFT (European Atrial Fibrillation Trial) Study Group. Lancet 1993;342:1255-1262.
Shapira Y, Vaturi M, Sagie A. Anticoagulant Albers GW, Dalen JE, Laupacis A, Manning WJ, Petersen P, Singer DE. Antithrombotic therapy in prosthetic valves undergoing non-cardiac surgery: atrial fibrillation. Chest 2001;119:194S-206S.
indications and unresolved issues. J Heart Valve Kopecky SL, Gersh BJ, McGoon MD, et al. The natural history of lone atrial fibrillation. A Spandorfer J. The management of anticoagulation population-based study over three decades. N before and after procedures. Med Clin North Am Douketis JD, Foster GA, Crowther MA, Prins MH, Ginsberg JS. Clinical risk factors and timing management and reversal of antithrombotic of recurrent venous thromboembolism during the therapy. Clin Ger Med 2001;17:189-201.
initial 3 months of anticoagulant therapy. Arch chronically anticoagulated patient. J Thromb Heit JA, Mohr DN, Petterson TM, O  Fallon WM, Melton LJ. Predictors of recurrence after deep Tiede DJ, Nishimura RA, Gastineau DA, Mullany vein thrombosis and pulmonary embolism. A population-based cohort study. Arch Intern Med.
management of prosthetic valve anticoagulation.
Schulman S, Svenungsson E, Granqvist S, and Husebye DG, Pluth JR, Piehler JM, et al.
the Duration of Anticoagulation Study Group.
Reoperation on prosthetic heart valves: an analysis of risk factors in 552 patients. J Thorac recurrence of thromboembolism and death among patients with venous thromboembolism following Martinelli J, Jiminez A, Rabago G, Artiz V, anticoagulant therapy. Am J Med 1998;104:332- Fraile J, Farre J. Mechanical carfiac valve Kearon C, Gent M, Hirsh J, Weitz J, Kovacs MJ, Circulation 1991;84 (Suppl III):70-75.
Anderson DR, et al. A comparison of three Vongpatanasin W, Hillis D, Lange RA. Prosthetic months of anticoagulant therapy with extended heart valves. N Engl J Med 1996;335:407-416.
anticoagulation for a first episode of idiopathic Baudet EM, Oca CC, Roques XF, et al. A 51/2 year experience with the St. Jude medical cardiac valve prosthesis: Early and late results of 737 Schulman S, Rhedin A-S, Lindmarker P, et al. A valve replacements in 671 patients. J Thorac comparison of six weeks with six months of oral anticoagulant therapy after a first episode of Cannagieter SC, Rosendaal FR, Briet E.
Thromboembolic and bleeding complications in patients with mechanical heart valve prostheses.
Research Committee of the British Thoracic Society. Optimum duration of anticoagulation for Atrial Fibrillation Investigators. Risk factors for deep-vein thrombosis and pulmonary embolism.
stroke and efficacy of anti-thrombotic therapy in atrial fibrillation: analysis of pooled data from the Levine MN, Hirsh J, Gent M, et al. Optimal five randomized controlled trials. Arch Intern duration of oral anticoagulant therapy: A randomized trial comparing four weeks with three Hart RG, Pearce LA, McBride R, et al. Factors months of warfarin in patients with proximal deep associated with ischemic stroke during aspirin vein thrombosis. Thromb Haemost. 1995;74:606- therapy in atrial fibrillation: analysis of 2012 participants in the SPAF I-III clinical trials: Ehrlichman RJ, Seckel BR, Bryan DJ, Moschella Stroke Prevention in Atrial Fibrillation (SPAF) CJ. Common complications of wound healing.
Prevention and management. Surg Clin North Am Field D. Assessment of haemostasis. Nurs Times Vinther CC, Settgast IW, Gram J. The effect of low-dose acetylsalicylic acid in bleeding after Nielson JD, Gram J, Holm-Nielsen A, et al. Post- transurethral prostatectomy - a prospective, randomized, double-blind, placebo-controlled prostatectomy is dependent on in situ fibrinolysis.
study. Scand J Urol Nephrol 2000;34:194-198.
Yende S, Wunderkind RG. Effect of clopidogrel Chakravarti A, MacDermott S. Transurethral on bleeding after coronary artery bypass surgery.
resection of the prostate in the anticoagulated Patrono C, Coller B, Dalen JE, et al. Platelet- Hergesell O, Felten H, Andrassy K, et al. Safety active drugs: the relationships among dose, of ultrasound-guided percutaneous renal biopsy - retrospective analysis of 1090 consecutive cases.
Nephrol Dial Transplant 1998;4:975-977.
Palareti G, Legnani C. Warfarin withdrawal.
Van Os EC, Kamath PS, Gostout CJ, Heit JA.
Gastroenterological procedures among patients considerations. Clin Pharmacokinet 1996;30:300- with disorders of hemostasis: evaluation and White RH, McKitrick T, Hutchison R, et al.
Temporary discontinuation of warfarin therapy: Jung S, Fehr S, Harder-d  Heureuse J, Widenmann changes in the international normalized ratio. Ann B, Dignass AU. Corticosteroids impair intestinal epithelial wound repair. Scand J Gastrointest Crowther MA, Julian J, McCarty D, Douketis J, Kovacs M, Schnurr T, et al. Low-dose oral Kadakia SC, Angueira CE, Ward JA, Moore M.
vitamin K for the treatment of warfarin-associated Gastrointestinal endoscopy in patients taking coagulopathy. Lancet 2000;356:1551-3.
antiplatelet agents and anticoagulants: survey of McGinnis, Ginsberg J. Low-dose oral vitamin K reliably reverses over-anticoagulation due to Despotis GJ, Skubas NJ, Goodnough LT.
warfarin. Thromb Haemost 1998;79:1116-1118.
Optimal management of bleeding and transfusion Anderson DR, Levine MN, Gent M, et al.
in patients undergoing cardiac surgery. Semin Validation of new criteria for the evaluation of Thorac Cardiovasc Surg 1999;2:84-104.
bleeding during antithrombotic therapy. Thromb Milas BL, Jobes DR, Gorman RC. Management of bleeding and coagulopathy after heart surgery.
Lausen I, Jensen R, Jorgensen LN, et al.
Semin Thorac Cardiovasc Surg 2000;4:326-36.
thrombosis occurring later after general surgery: thromboprophylaxis. Eur J Surg 1998;164:657- Landefeld S, Cook F, Flately M, et al.
Identification and preliminary validation of Dickinson LD, Miller LD, Patel CP, et al.
predictors of major bleeding in hospitalized patients starting anticoagulant therapy. Am J Med postoperative intracranial hemorrhage in when Despotis GJ, Filos KS, Zoys TN, Hogue CW Jr, thrombosis prophylaxis in patients with brain Spitznagel E, Lappas DG. Factors associated with tumors. Neurosurgery 1998;43:1074-1081.
Dillioglugil O, Liebman BD, Leibman NS, et al.
Risk factors for complications and morbidity after multivariate analysis in cardiac surgical patients.
radical prostatectomy. J Urol 1997;157:1760- Kallis P, Tooze JA, Talbot S, Cowans D, Bevan DH, Treasure T. Pre-operative aspirin decreases Kuwada SK, Balm R, Gostout CJ. The risk of platelet aggregation and increases post-operative withdrawing chronic anticoagulation because of blood loss - a prospective, randomized, placebo- controlled, double-blind clinical trial in 100 patients with chronic stable angina. Eur J Ananthasubramanian K, Beattie JN, Rosman HS, Jayam V, Borzak S. How safely and for how long catheter removal in patients receiving twice-daily can warfarin therapy be withheld in prosthetic or once-daily low-molecular-weight heparin and heart valve patients hospitalized with major continuous epidural analgesia after orthopedic surgery. Thromb Haemost 2002 (in press). Gomez CR, Sandhu J, Mehta P. Resumption ofanticoagulation during hypertensive cerebralhemorrhage with prosthetic heart valve [letter].
Stroke 1988;19:407.
Babikian VL, Kase CS, Pessin MS, et al.
Resumption of anticoagulation after intracranialbleeding in patients with prosthetic heart valves[letter]. Stroke 1988;19:407-408.
Kawamata T, Takeshita M, Kubo O, et al.
Management of intracranial hemorrhageassociated with anticoagulant therapy. SurgNeurol 1995;44:438-443.
Cruickshank MK, Levine MN, Hirsh J, et al. Astandard heparin nomogram for the managementof heparin therapy. Arch Intern Med1991;151:333-337.
Morabia A. Heparin doses and major bleeding.
Lancet 1986;1:1278-1279.
Hollingsworth JA, Rowe BH, Brisebois FJ, et al.
The successful application of a heparinnomogram in a community hospital. Arch InternMed 1995;155:2095-2100.
Prandoni P, Bagatella P, Bernardi E, et al. Use ofan algorithm for administering subcutaneousheparin in the treatment of deep venousthrombosis. Ann Intern Med 1998;129:299-302.
Cadroy Y, Pourrat Y, Baladre MF, et al. Delayedelimination of enoxaparin in patients with chronicrenal insufficiency. Thromb Res 1991;63:385-390.
Warkentin TE. Heparin-inducedthrombocytopenia: a clinicopathologic syndrome.
Thromb Haemost 1999;82 (suppl):439-447.
Greinacher A, Janssens U, Berg G, Bock M,Kwasny H, Kemkes-Matthes B, et al. Lepirudin(recombinant hirudin) for parenteralanticoagulation in patients with heparin-inducedthrombocytopenia. Heparin-AssociatedThrombocytopenia Study (HAT) investigators.
Circulation 1999;100:587-593.
Horlocker TT. Low molecular weight heparin andneuraxial anesthesia. Thromb Res2001; 101: V141-154.
Lawton MT, Porter RW, Heiserman JE,Jacobowitz R, Sonntag VKH, Dickman CA.
Surgical management of spinal epiduralhematoma: relationship between surgical timingand neurological outcome. J Neurosurg 1995; 83:1-7.
Douketis JD, Kinnon K, Crowther MA.
Anticoagulant effect at the time of epidural Thromboembolic Events in Patients with a Mechanical Heart Valve after
Temporary Interruption of Warfarin Therapy
Thromboembolic Events/Patients at Risk (percent) *Intravenous UFH (unspecified dose); **no perioperative warfarin withdrawal; subcutaneous or intravenous UFH (dose unspecified) or no warfarin withdrawal;enoxaparin 1 mg/kg twice-daily; dalteparin 100 IU/kg twice-daily; anticoagulants withheld because of bleeding episode.
Perioperative Anticoagulant Management in Patients with a Mechanical
Prosthetic Heart Valve
- any mitral valve- caged-ball* or tilting-disc aortic valve - bileaflet aortic valve and <2 stroke risk *Starr-Edwards valve; Bjork-Shiley, Medtronic-Hall or Omnicarbon valve; St.Jude or Carbomedics valve; stroke risk factors: atrial fibrillation, previous stroke, transient ischemic attack or systemic embolism, left ventricular dysfunction,age >75 years, hypertension, diabetes mellitus.
Risk Factors for Stroke in Patients with Chronic Atrial Fibrillation
*Ref. 38; ref. 39; statistically significant in univariate analysis only; n/a, not Perioperative Anticoagulant Management in Patients with Chronic Atrial
Fibrillation
- rheumatic mitral valvular heart disease - chronic atrial fibrillation and 2 or more - chronic atrial fibrillation and <2 stroke *Stroke risk factors: previous stroke, transient ischemic attack or systemicembolism, left ventricular dysfunction, age >75 years, hypertension, diabetesmellitus.
Perioperative Anticoagulant Management in Patients with Venous
Thromboembolism
- antiphospholipid antibody- major comorbid disease  previous interruption of warfarin therapy bridging anticoagulant therapy is optional VTE = venous thromboembolism; *cancer that has been treated within 6 monthsor is palliative; anticardiolipin antibody or lupus anticoagulant; chronic cardiac Bleeding Risk Classification and Postoperative Anticoagulant Management
- coronary artery bypass graftsurgery- cervical cone biopsy- renal biopsy- bowel polypectomy - major orthopedic surgery- multiple dental extractions - laparoscopic cholecystectomy orhernia repair- single dental extraction *Low-dose LMWH (see Table 7) is the initial dose administered, but can be
continued for 24 to 72 hours in patients at increased risk of bleeding until the start
of full-dose LMWH; see Table 8.
Initial Low-Dose LMWH Therapy after Surgery or Invasive Procedure
*Administered as a single dose on the evening of the day of surgery, or for anadditional 24 to 72 hours in patients at increased risk of bleeding; IU, internationalunits.
Full-Dose LMWH Therapy after Surgery or Invasive Procedure*
*Full-dose (therapeutic-dose) LMWH to follow initial treatment with low-doseLMWH; IU, international units.
Suggested Perioperative Anticoagulation Patient Care Path
STOP warfarin (i.e., no warfarin on this day) LMWH _____________units, once-daily (given 24 hours before surgery) LMWH ____________units (evening dose, when hemostasis adequate)warfarin _______ mg LMWH ____________units, once-daily (24 hours after surgery, whenhemostasis adequate)warfarin _______ mg LMWH _____________units, once-dailywarfarin _______ mg LMWH ____________units, once-daily (if required)warfarin _______ mg LMWH ____________units, once-daily (if required) *The dose of LWMH will depend on the bleeding risk associated with surgery
(Table 4); the first dose of LMWH should be low-dose (Table 5); subsequent
LMWH administration can be increased to full-dose (Table 6).

Source: http://uk.oocities.com/mim_thrombo/douketis-review.pdf