Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
ORIGINAL INVESTIGATION Low-Molecular-Weight Heparin as Bridging Anticoagulation During Interruption of Warfarin Assessment of a Standardized Periprocedural Anticoagulation Regimen James D. Douketis, MD, FRCPC; Judith A. Johnson, RN; Alexander G. Turpie, MB, FRCPC Background: The treatment of patients at increased risk for arterial thromboembolism who require temporary interruption of warfarin sodium therapy is a common clinical problem. We investigated the efficacy and safety of a standardized periprocedural anticoagulation regimen with low-molecular-weight heparin. Methods: We studied 650 consecutive patients with a mechanical heart valve, chronic atrial fibrillation, or embolic stroke who required interruption of warfarin therapy because of an invasive procedure. Warfarin was stopped 5 or 6 days before the procedure, and patients received subcutaneous dalteparin sodium, 100 IU/kg twice daily, starting 3 days before the procedure. The risk of postprocedural bleeding determined postprocedural anticoagulant management. In patients undergoing a non– high-bleeding-risk procedure who had adequate postprocedural hemostasis, warfarin was resumed on the evening of the procedure, and dalteparin sodium, 100 IU/kg twice daily, was resumed on the next day and continued until the international normalized ratio was 2.0 or more. If postprocedural hemostasis was not secured, the resumption of dalteparin was delayed. In patients undergoing a high-bleeding-risk procedure, warfarin was From the Departments of Medicine, McMaster University (Drs Douketis and Turpie and Ms Johnson) and St Joseph’s Hospital (Dr Douketis), and Hamilton Health Sciences, General Hospital (Ms Johnson and Dr Turpie), Hamilton, Ontario. Drs Douketis and Turpie have received honoraria for speaking engagements and participation in advisory committees from companies that make low-molecularweight heparin. resumed on the evening of the procedure, but dalteparin was not given after the procedure. Results: Patients were followed up during the preprocedural and postprocedural period for a mean of 13.8 days (range, 10-18 days). In 542 patients who underwent a non–high-bleeding-risk procedure, there were 2 thromboembolic events (0.4%), 4 major bleeding episodes (0.7%), and 32 episodes of increased wound-related blood loss that precluded postprocedural dalteparin administration (5.9%). In 108 patients who underwent a highbleeding-risk procedure, there were 2 deaths (1.8%) possibly due to thromboembolism and 2 major bleeding episodes (1.8%). Conclusions: In patients at increased risk for arterial thromboembolism who require temporary interruption of warfarin therapy, a standardized periprocedural anticoagulant regimen with low-molecular-weight heparin is associated with a low risk of thromboembolic and major bleeding complications. Arch Intern Med. 2004;164:1319-1326 T HE TREATMENT OF PAtients with a mechanical heart valve or chronic atrial fibrillation who require temporary interruption of warfarin sodium therapy because of surgery or another invasive procedure is a frequently encountered but underinvestigated clinical problem.1-4 A major gap in knowledge is a lack of reliable estimates as to the incidence of thromboembolic events associated with warfarin therapy interruption.1 It is well established, however, that such events can have devastating clinical consequences: thrombosis of a mechanical heart valve is fatal in 15% of patients,5,6 and embolic stroke results in a major neurologic deficit or death in 70% of patients.7 Consequently, despite disagreement on the optimal periprocedural (REPRINTED) ARCH INTERN MED/ VOL 164, JUNE 28, 2004 1319 anticoagulation strategy during interruption of warfarin therapy,8-12 several authorities1-4 and consensus groups13,14 advocate, for most patients, some form of bridging therapy with a short-acting anticoagulant. The rationale for bridging anticoagulation is to minimize the time before and after a procedure that patients are not receiving therapeutic anticoagulation and, thereby, minimize the risk of thromboembolism. The conventional periprocedural anticoagulation approach is to hospitalize patients 4 to 5 days before surgery to stop warfarin and administer intravenous unfractionated heparin while the anticoagulant effect of warfarin recedes.15,16 Intravenous heparin is stopped 3 to 4 hours before the procedure to avoid a residual anticoagulant effect at the time of the procedure. After the procedure, war- WWW.ARCHINTERNMED.COM ©2004 American Medical Association. All rights reserved. METHODS Table 1. Periprocedural Anticoagulation Regimen Day −7 to −10 −7 −5 or −6 −3 or −4 −1 0 Preprocedure Assess for perioperative bridging anticoagulation, classify patients as undergoing high-bleeding-risk or non–high-bleeding-risk procedure Stop aspirin (or other antiplatelet drugs)* Stop warfarin sodium† Start subcutaneous dalteparin sodium, 100 IU/kg, twice daily Last preprocedural dose of dalteparin administered not less than 12 h before start of surgery or procedure Day of Procedure Assess postoperative hemostasis after surgery or procedure; for non–high- or high-bleeding-risk procedures, resume warfarin usually on evening of or day after procedure‡ Postprocedure Non–High-Bleeding-Risk Procedure +1 +4 +7 to +10 PATIENT REGISTRY Intervention Start dalteparin sodium, 100 IU/kg, twice daily§ INR testing (stop dalteparin if INR ⱖ2.0) INR testing High-Bleeding-Risk Procedure No dalteparin administration INR testing Starting in November 1997, a structured clinical management protocol for patients who required temporary interruption of warfarin therapy was initiated at the Hamilton Health Sciences, General Hospital, a tertiary care teaching hospital and regional center for cardiovascular and neurologic diseases in Hamilton, Ontario. Patients received a standardized regimen of periprocedural bridging anticoagulation with LMWH that was administered, whenever feasible, out-of-hospital by the patient or another caregiver. This management protocol was developed in response to an increasing number of elective surgical and other invasive procedures that were being undertaken without hospitalization in patients receiving warfarin, and because of limited hospital bed availability that precluded periprocedural intravenous heparin administration. To assess the efficacy and safety of this novel periprocedural anticoagulation strategy and to allow internal auditing of patient quality of care, we established a prospective registry of patients who received bridging anticoagulation with LMWH. Because this periprocedural anticoagulation regimen became the standard of care at our institution, informed consent was not required for inclusion into this patient registry. PATIENTS INR testing Abbreviation: INR, international normalized ratio. *After procedure, resumed on the same day as warfarin resumption. †Warfarin stopped on day −5 if target INR is 2.0 to 3.0 and on day −6 if target INR is 2.5 to 3.5. ‡Warfarin resumed when the patient is able to take medications by mouth. §Start of dalteparin may be delayed until hemostasis is secured. farin and intravenous heparin are resumed, the latter administered for 4 to 5 days until therapeutic anticoagulation with warfarin is reestablished. This approach is difficult to implement because of the current constraints on hospital bed availability and the increasing number of surgical and other invasive procedures that are being performed without hospitalization. An alternative periprocedural management strategy is the use of low-molecular-weight heparin (LMWH) for bridging anticoagulation. This approach is appealing because LMWH can be administered subcutaneously, in a fixed weight-based dose, and without the need for laboratory monitoring,17 thereby obviating the need for hospitalization to administer anticoagulants. Furthermore, this strategy has the potential to substantially reduce health care costs.18 However, there is little evidence to support the efficacy and safety of LMWH as bridging anticoagulation. Previous studies of periprocedural anticoagulation are limited because no standardized anticoagulant regimen was investigated,19-22 anticoagulants other than LMWH were investigated,19-23 or studies of LMWH as bridging therapy involved fewer than 30 patients.24-26 We report on a prospective registry of 650 patients at risk for arterial thromboembolism who required temporary interruption of warfarin therapy and received bridging anticoagulation with LMWH. This is the first largescale study, to our knowledge, assessing the efficacy and safety of a standardized periprocedural anticoagulation regimen with LMWH. Consecutive adult patients, 18 years or older, at risk for arterial thromboembolism who were assessed in the hospital Anticoagulation Clinic between November 1, 1997, and June 30, 2002, for temporary interruption of warfarin therapy were eligible for this registry. Patients satisfied the following criteria for inclusion: (1) receiving warfarin therapy, with a target international normalized ratio (INR) of 2.0 to 3.5; (2) mechanical heart valve, chronic atrial fibrillation, or a previous stroke or transient ischemic attack with a presumed embolic source; and (3) undergoing an elective surgical or other invasive procedure that requires normalization of the INR. Patients were excluded from the registry if one or more of the following characteristics were present: (1) renal insufficiency (serum creatinine level ⬎2.0 mg/dL [⬎178 mmol/L]) that precluded therapeutic-dose LMWH administration; (2) previous heparininduced thrombocytopenia; (3) pregnancy; (4) treatment with an anticoagulant other than the prespecified LMWH regimen; or (5) undergoing a minor dental procedure, such as teeth cleaning or single-tooth extraction, in which reversal of anticoagulation is not required.27 Patients who were to receive spinal anesthesia were eligible for this registry but were excluded if they had an indwelling epidural catheter after the procedure for analgesia, which precluded coadministered LMWH.28 Patients who required rapid reversal of anticoagulation because of an urgent procedure were not included in this registry. PERIPROCEDURAL ANTICOAGULANT MANAGEMENT Patients received a standardized anticoagulation regimen, as outlined in Table 1, that involved therapeutic-dose LMWH as bridging therapy because less intense regimens, with lowdose LMWH, had not been previously assessed for efficacy. Patients were instructed in subcutaneous self-injection of LMWH by the Anticoagulation Clinic nurse practitioner, and syringes were prefilled to minimize dose errors. Patients were provided with information about potential bleeding and thromboembolic complications, and were asked to report such events if they occurred or were suspected. A 24-hour telephone con- (REPRINTED) ARCH INTERN MED/ VOL 164, JUNE 28, 2004 1320 WWW.ARCHINTERNMED.COM ©2004 American Medical Association. All rights reserved. tact number was provided if patients had difficulty with the LMWH injections or other health-related concerns. If a patient or caregiver was unable to administer LMWH, it was undertaken by a visiting nurse. For preprocedural anticoagulation, in general, warfarin therapy was interrupted on the fifth day before the procedure in patients receiving warfarin with a target INR of 2.0 to 3.0, and on the sixth day before the procedure in patients receiving warfarin with a target INR of 2.5 to 3.5. The INR testing was done on the third or fourth day before the procedure. If the INR was less than 2.5, treatment was initiated with dalteparin sodium, 100 IU/kg twice daily by subcutaneous injection, with the first dose supervised by the Anticoagulation Clinic nurse. If the INR was 2.5 or more, thereby precluding the initiation of dalteparin therapy, an injection of isotonic sodium chloride solution was administered subcutaneously to observe the patient’s self-injection technique, and the first dose of dalteparin was given on the next day. To minimize the likelihood of a residual anticoagulant effect at the time of the procedure, the last preprocedure dalteparin dose was administered not less than 12 hours before the start of the procedure, and if the INR was 3.0 or more on the third or fourth day before the procedure, patients received phytonadione, 1 mg orally. If patients were receiving antiplatelet therapy, usually aspirin, this was interrupted 7 days before the procedure. The postprocedural anticoagulation regimen was determined on the basis of 2 considerations: (1) the bleeding risk associated with the procedure and (2) the adequacy of postprocedural hemostasis. Patients were classified as being at high risk or nonhigh risk of postprocedural bleeding. This classification of bleeding risk was based on a subjective assessment of the anticipated procedure-related bleeding, which was determined by input from the anticoagulation physician and nursepractitioner and the attending surgeon or interventionist. The adequacy of postprocedural hemostasis was determined by the same caregivers and was based on a subjective assessment of wound-related blood loss. In patients undergoing a high-bleeding-risk procedure, warfarin therapy was resumed on the evening after the procedure, but patients did not receive dalteparin at any time after the procedure. In patients undergoing a non–high-bleeding-risk procedure who had adequate postprocedural hemostasis, warfarin was resumed on the evening of procedure, and dalteparin sodium, 100 IU/kg twice daily, was resumed on the day after the procedure, with the first dose administered approximately 24 hours after the procedure. Dalteparin was continued until the INR was 2.0 or more. In patients undergoing a non–high-bleedingrisk procedure who had inadequate postprocedural hemostasis, the resumption of warfarin was delayed until the first postprocedural day, and dalteparin was delayed until the second or third postprocedural day when hemostasis was secured. In all patients, the initial postprocedural dose of warfarin corresponded to the patient’s usual dose for that day of the week, and antiplatelet therapy was resumed on the same day as dalteparin. CLINICAL OUTCOMES AND PATIENT FOLLOW-UP Three primary clinical outcomes were assessed in all patients: (1) thromboembolism, (2) major bleeding, and (3) death. Thromboembolism was defined as a stroke or transient ischemic attack, an acute coronary syndrome, or systemic embolism involving a limb or viscera that was ascertained on the basis of clinical findings and objective diagnostic testing. Major (or clinically important) bleeding was defined as bleeding that was clinically overt, with new symptoms (eg, hematemesis) or signs (eg, wound hematoma), and was associated with one or more of the following characteristics: hemoglobin level decrease of more than 20 g/L over 24 hours; transfusion of 2 U or more of packed red blood cells; or bleeding at a critical site (retroperitoneal, intracranial, or body cavity).29 Patients who had a decrease in hemoglobin level or received a blood transfusion because of expected procedure-related blood loss, but with no clinically overt bleeding, were not considered to have major bleeding. Death was documented on the basis of information from patients’ medical records or death certificate, and if it occurred suddenly, it was classified as possibly due to thromboembolism. A secondary clinical outcome was increased woundrelated blood loss, which was assessed only in patients undergoing non–high-bleeding-risk procedure in whom the presence of increased blood loss would delay or suspend the postprocedural resumption of dalteparin therapy. This subjective outcome was characterized by greater-than-expected oozing from a wound site or blood accumulation in a wound drain. Patients underwent follow-up from the time warfarin therapy was interrupted before the procedure until therapeutic anticoagulation with warfarin, defined by an INR 2.0 or more, was reestablished after the procedure. Subsequent warfarin dosing was undertaken by the Anticoagulation Clinic or the patient’s primary care physician. STATISTICAL ANALYSIS Statistical analyses were done with SAS 8.2 software (SAS Institute Inc, Cary, NC). Baseline patient characteristics and periprocedural anticoagulation regimens were expressed as a mean with a corresponding standard deviation. Clinical outcome rates were based on the number of patients with a thromboembolic or bleeding event divided by the number of patients at risk and were expressed as a proportion with a corresponding 95% confidence interval. RESULTS PATIENTS Of 798 patients at risk for arterial thromboembolism who were assessed for temporary interruption of warfarin because of an elective surgical or other invasive procedure, 104 were excluded from the registry because the patient was undergoing a minor dental procedure (n=89), had renal insufficiency (n=14), or had previous heparininduced thrombocytopenia (n = 1). Of 694 registry patients who were scheduled to receive the prespecified anticoagulation regimen, 35 were excluded because they received an LMWH other than dalteparin before or after the procedure, and 9 were excluded because they received intravenous heparin after the procedure at the discretion of the treating physician (Figure). Thus, there were 650 patients scheduled to receive the prespecified anticoagulation regimen, of whom 542 (83%) were undergoing a non–high-bleeding-risk procedure, and 108 (17%) were undergoing a high-bleeding-risk procedure. The baseline patient characteristics are described in Table 2. The type of procedure patients underwent and the associated bleeding risk category are described in Table 3. All patients received out-of-hospital dalteparin injections before the procedure, which was administered by the patient (self-injection), a family member, or a visiting nurse in 88%, 7%, and 5% of cases, respectively. After the procedure, dalteparin was administered by the patient, a family member, or a visiting nurse in 85%, 7%, and 8% of cases, respectively. (REPRINTED) ARCH INTERN MED/ VOL 164, JUNE 28, 2004 1321 WWW.ARCHINTERNMED.COM ©2004 American Medical Association. All rights reserved. 798 Patients Assessed for Bridging Anticoagulation 694 Patients Scheduled to Receive Prespecified Periprocedural Bridging Anticoagulation With Dalteparin Sodium, 100 IU/kg SC BID 104 Patients Excluded From Registry: Minor Dental Procedure (89) Renal Insufficiency (14) Heparin-Induced Thrombocytopenia (1) 44 Patients Excluded From Analysis: Received Intravenous Heparin (9) Received Different LMWH (35) Stratification of Patients According to Bleeding Risk Associated With Procedure 542 Patients Underwent Non–High-Bleeding-Risk Procedure (Scheduled to Receive Preprocedural and Postprocedural Dalteparin Sodium, 100 IU/kg SC BID) 108 Patients Underwent High-Bleeding-Risk Procedure (Scheduled to Receive Only Preprocedural Dalteparin Sodium, 100 IU/kg SC BID) Resumption of Dalteparin: 454 Patients on Day + 1 After Procedure 46 Patients on Day + 2 After Procedure 5 Patients on Day + 3 After Procedure 5 Patients on Day + 4 After Procedure Suspension of Dalterparin: 32 Patients Did Not Receive Postprocedural Dalteparin (Increased Wound-Related Blood Loss) Perioperative anticoagulant management. SC indicates subcutaneously; BID, twice daily; and LMWH, low-molecular-weight heparin. For preprocedural anticoagulation, warfarin therapy was interrupted a mean of 5.7 days before the procedure, and patients received a mean of 5.4 doses of dalteparin. For postprocedural anticoagulation, in 108 patients who underwent a high-bleeding-risk procedure and were not scheduled to receive only postprocedural dalteparin, warfarin was resumed on the evening of or the day after the procedure. In the 542 patients who underwent a non–high-bleeding-risk procedure, there was excessive postprocedural bleeding in 32 patients that precluded postprocedural dalteparin administration, and these patients received only warfarin, starting the day after the procedure. In 510 patients who underwent a non– high-bleeding-risk procedure and received postprocedural dalteparin, it was resumed on the first postprocedure day in 454 patients (89%), at least 24 hours after the procedure, on the second postprocedural day in 46 patients (9%), on the third postprocedural day in 5 patients (1%), and on the fourth postprocedural day in 5 patients (1%). This delay in resumption of postprocedural dalteparin in 56 patients was necessitated by delayed hemostasis (23 patients) and concerns about bleeding after implantation of a permanent pacemaker or internal cardiac defibrillator (30 patients) or bowel polypectomy (3 patients). PATIENT FOLLOW-UP AND ADVERSE CLINICAL OUTCOMES Table 2. Characteristics of 650 Registry Patients Characteristic Finding Mean age (range), y Mean weight (range), kg Sex, No. (%) Men Women Reason for warfarin sodium therapy, No. (%) AF AF and valvular heart disease AF and cardiomyopathy 67.8 (28-88) 80.2 (55-99) 417 (64) 233 (36) 306 (47) 22 (3) 18 (3) Any AF 346 (53) Mechanical AVR Mechanical AVR and Bental procedure Mechanical AVR and AF Mechanical MVR Mechanical MVR and AF Mechanical AVR and MVR 108 (17) 2 (⬍1) 33 (5) 29 (4) 17 (3) 26 (4) Any mechanical heart valve 215 (33) Previous stroke/TIA and AF Previous stroke/TIA and mechanical AVR Previous stroke/TIA and mechanical MVR Previous stroke/TIA and mechanical AVR and MVR 70 (11) 9 (1) 6 (1) 4 (⬍1) Any previous stroke 89 (14) Abbreviations: AF, atrial fibrillation, AVR, aortic valve replacement; MVR, mitral valve replacement; TIA, transient ischemic attack. PERIPROCEDURAL ANTICOAGULANT MANAGEMENT The periprocedural anticoagulation dosing regimen patients received is outlined in Table 4 according to patients’ indications for warfarin therapy and their corresponding target INR. Patients underwent clinical follow-up during the preprocedural and postprocedural period for a mean of 13.8 days (range, 10-18 days). The rates of adverse clinical outcomes are outlined in Table 5. In 108 patients undergoing a high-bleeding-risk procedure who received only preprocedural dalteparin, there were 2 nonfatal major bleeding episodes (1.8%) (upper gastrointestinal tract bleeding and wound hematoma), neither of which was fatal; no confirmed thromboembolic events; and 2 deaths (1.8%) (cardiac arrests) possibly due to thromboembolism. In 542 patients undergoing a non–high-bleedingrisk procedure who were scheduled to received preprocedural and postprocedural dalteparin, there were 4 major bleeding episodes (0.7%) (3 wound hematomas and a rectus abdominus sheath hematoma), none of which were fatal; 2 thromboembolic events (0.4%) (systemic embolism and transient ischemic attack); and 32 episodes (5.9%) of excessive wound-related bleeding that precluded postprocedural administration of dalteparin. The adverse clinical outcomes, which all occurred during the postprocedural period, are described in Table 6. COMMENT In this registry of 650 patients at increased risk for arterial thromboembolism who required temporary interruption of warfarin therapy for an elective procedure, a standardized bridging anticoagulation regimen with LMWH was associated with a low risk of thromboembolic (⬍1%) and major bleeding (1%-2%) complications in the immediate periprocedural period, and was feasible for out-of-hospital administration. This anticoagulation strategy involved 2 main components: (1) preprocedural and postprocedural dalteparin sodium therapy, (REPRINTED) ARCH INTERN MED/ VOL 164, JUNE 28, 2004 1322 WWW.ARCHINTERNMED.COM ©2004 American Medical Association. All rights reserved. Table 3. Classification of Bleeding Risk Associated With Procedure Bleeding Risk Category High Surgical or Other Invasive Procedure Heart valve replacement ± coronary artery bypass Coronary artery bypass Abdominal aortic aneurysm repair Neurosurgical cancer surgery Urogynecologic cancer surgery Head and neck cancer surgery Intra-abdominal and breast cancer surgery Bilateral knee replacement Laminectomy Transurethral prostate resection Kidney biopsy Total Nonhigh (surgical procedure) Total Nonhigh (nonsurgical procedure) Cholecystectomy Bowel resection Abdominal hernia repair Hemorrhoidal surgery Bowel polypectomy Axillary node dissection Abdominal hysterectomy Dilation and curettage Hydrocele repair Hand surgery Carpal tunnel repair Skin cancer excision Knee replacement Hip replacement Shoulder surgery Foot surgery Pacemaker insertion Sternotomy wire removal Internal cardiac defibrillator insertion Endarterectomy or carotid bypass surgery Noncataract eye surgery Cataract eye surgery Dental surgery Coronary angiography ± percutaneous coronary intervention Electrophysiologic testing Noncoronary angiography Gastrointestinal endoscopy ± biopsy Bronchoscopy ± biopsy Arthroscopy Central venous catheter removal Biopsy (prostate, bladder, thyroid, breast, lymph node, pancreas, myocardial thyroid) Total 100 IU/kg twice daily, to minimize the duration without therapeutic anticoagulation; and (2) avoidance of postprocedural dalteparin in patients at high risk of bleeding or with inadequate postprocedural hemostasis, and delay in dalteparin resumption in patients with delayed hemostasis. The second component was of paramount importance because the development of postprocedural major bleeding would delay the resumption of anticoagulation and, thereby, increase the time that patients would be exposed to the risk of thromboembolism. This patient registry has several features that support the validity of the results. First, unlike most registries in which patient treatment is left to the discretion of the treating physician,30,31 patients in this registry received a standardized periprocedural anticoagulation regimen. This would reduce the potential for variability in the aggressiveness of periprocedural anticoagula- No. of Procedures (%) 61 6 5 2 3 4 5 2 9 10 1 108 (17) 17 6 21 4 5 1 2 3 1 6 4 16 4 2 4 5 56 4 11 9 6 17 3 205 (31) 223 11 8 64 3 7 1 21 337 (52) tion8,11,12 or cointerventions such as periprocedural aspirin therapy that could influence the risk of thromboembolic and bleeding events.32,33 Second, patients underwent a standardized duration of clinical followup, thereby limiting variability in patient surveillance that would influence the likelihood of capturing clinical outcomes. Third, unlike most registries that allow broad patient eligibility criteria, this registry assessed a prespecified patient group at increased risk for arterial thromboembolism, in whom most physicians would consider some form of periprocedural bridging anticoagulation.1-4,13,14 Clinically important (or major) bleeding complications were infrequent, occurring in 1% to 2% of patients. The low incidence of major bleeding is likely attributable to 3 factors. First, 52% of patients underwent nonsurgical invasive procedures, such as cardiac cath- (REPRINTED) ARCH INTERN MED/ VOL 164, JUNE 28, 2004 1323 WWW.ARCHINTERNMED.COM ©2004 American Medical Association. All rights reserved. Table 4. Periprocedural Anticoagulation Regimen Mean ± SD Patient Group (Target INR) Atrial fibrillation (2.0-3.0) Mechanical heart valve (2.5-3.5) Previous stroke or TIA (2.0-3.0) Anticoagulation Management Before Procedure After Procedure Warfarin stopped, d INR when warfarin stopped Dalteparin sodium injections, No. First dalteparin sodium dose, IU Warfarin stopped, d INR at time of stopping Dalteparin sodium injections, No. First dalteparin sodium dose, IU Warfarin stopped, d INR when warfarin stopped Dalteparin sodium injections, No. First dalteparin sodium dose, IU 5.6 ± 1.4 2.1 ± 0.6 5.4 ± 1.9 8071 ± 1781 6.2 ± 1.6 6.2 ± 1.6 5.1 ± 1.7 7767 ± 1468 5.8 ± 1.3 2.1 ± 0.7 4.9 ± 1.3 7523 ± 1788 NA NA 5.1 ± 1.1 8055 ± 1763 NA NA 5.1 ± 1.1 7782 ± 1459 NA NA 5.1 ± 1.1 7523 ± 1788 Abbreviations: INR, international normalized ratio; NA, not applicable; TIA, transient ischemic attack. Table 5. Adverse Clinical Outcome Rates Patient Group, % (95% CI) Clinical Outcome* Thromboembolism Thromboembolism (including possible events) Major bleeding Increased wound-related blood loss§ Patients Undergoing Non–High-Bleeding-Risk Procedure (n = 542)† Patients Undergoing High-Bleeding-Risk Procedure (n = 108)‡ All Patients (n = 650) 0.37 (0.04-1.32) 0.74 (0.20-1.87) 0.74 (0.20-1.87) 5.90 (4.07-8.23) None 1.85 (0.23-6.52) 1.85 (0.23-6.52) NA 0.31 (0.04-1.00) 0.62 (0.17-1.57) 0.92 (0.34-2.00) NA Abbreviations: CI, confidence interval; NA, not applicable. *All clinical outcomes occurred after the procedure. †Scheduled to receive preprocedural and postprocedural bridging anticoagulation with dalteparin sodium. ‡Scheduled to receive only preprocedural bridging anticoagulation with dalteparin. §Precluded scheduled postprocedural administration of dalteparin. eterization or gastrointestinal endoscopy, which are associated with a low risk of bleeding complications.34,35 Second, measures were taken to minimize the risk of bleeding. Postprocedural dalteparin was avoided in patients undergoing a high-bleeding-risk procedure such as open heart surgery,36 abdominal vascular surgery,37 neurosurgery,38 major cancer surgery,39,40 or procedures involving the prostate or kidney.41-43 In patients undergoing a non–high-bleeding-risk procedure, the start of dalteparin was delayed for 24 hours until the day after the procedure, when there was adequate wound hemostasis. This was done because in patients who receive LMWH after a surgical procedure, more than 90% of major bleeding episodes occur at the surgical wound site,44-46 and such bleeding is more likely to occur with early postprocedural initiation of anticoagulant therapy, 4 to 12 hours after surgery, than delayed initiation of anticoagulation, more than 12 hours after surgery.23,47 Furthermore, the start of dalteparin was delayed or suspended in patients with inadequate postprocedural hemostasis, and in those undergoing pacemaker or cardiac defibrillator implantation or bowel polypectomy, who might be at increased risk of bleeding complications with early postprocedural initiation of anticoagulation.48,49 Third, our definition of major bleeding was narrower than that used in other studies involving postoperative anticoagula- tion,44,47 as it included only clinically overt bleeding. There is a need for a bleeding classification scheme specific for the postprocedural clinical setting that differentiates between clinically important bleeding that requires a change in clinical management (eg, reoperation) and increased wound-related blood loss that is expected and may require transfusion but is, otherwise, self-limiting and without clinical consequences.50 Thromboembolic complications were also infrequent. There was 1 episode of systemic embolism to the lower extremities, 1 transient ischemic attack and 2 deaths, possibly due to thromboembolism, yielding an overall incidence of 0.6%. Reliable estimates of periprocedural thromboembolic complications during temporary interruption of warfarin treatment are limited. It is postulated that the absolute risk of thromboembolic events during the 2- to 4-day period after warfarin interruption, when there is no anticoagulant effect present, is a small fraction of the 5% to 15% annual risk of thromboembolic events.8 However, thromboembolic risk may be higher than expected because after interruption of warfarin therapy a transient hypercoagulable state may develop owing to a rebound increase in thrombin generation or platelet activation.51,52 Furthermore, a surgical procedure may induce a hypercoagulable state through mechanisms that include vessel-wall injury and fibrinolysis (REPRINTED) ARCH INTERN MED/ VOL 164, JUNE 28, 2004 1324 WWW.ARCHINTERNMED.COM ©2004 American Medical Association. All rights reserved. Table 6. Adverse Clinical Outcomes Patient Age, y/ Sex Reason for Warfarin Sodium Therapy 72/M Chronic AF Gastrointestinal endoscopy 82/F Chronic AF Cataract removal 72/F Chronic AF Pacemaker insertion 62/M Previous embolic stroke Abdominal hernia repair 68/M Chronic AF Coronary angiography 76/F MVR, chronic AF Dental surgery 72/F Chronic AF, previous TIA Bilateral knee replacement 74/F 75/F 70/M Previous embolic stroke AVR, MVR, chronic AF Chronic AF, valvular heart disease Intra-abdominal cancer surgery Lumbar laminectomy Heart valve replacement, coronary bypass Procedure Periprocedural Anticoagulation Clinical Outcome (Days After Procedure) Preprocedural and postprocedural dalteparin sodium Preprocedural and postprocedural dalteparin Preprocedural and postprocedural dalteparin Preprocedural and postprocedural dalteparin Preprocedural and postprocedural dalteparin Preprocedural and postprocedural dalteparin Preprocedural dalteparin Stroke, lower-limb embolism (5) Preprocedural dalteparin Preprocedural dalteparin Preprocedural dalteparin TIA (8) Wound hematoma (2) Wound hematoma (3) Wound hematoma (2) Rectus abdominus sheath hematoma (3) Upper gastrointestinal tract bleeding (5) Wound hematoma (2) Cardiac arrest (6) Cardiac arrest (7) Abbreviations: AF, atrial fibrillation; AVR, aortic valve replacement; MVR, mitral valve replacement; TIA, transient ischemic attack. inhibition.53-55 In 3 prospective studies that have investigated periprocedural bridging anticoagulation with LMWH,25,26,56 thromboembolic events occurred in 1 (0.8%) of 128 of patients, an event rate that is consistent with our findings. There are limitations of this study that should be addressed. First, we acknowledge that postprocedural patient follow-up was limited to about 1 week in most patients. Consequently, our findings may underestimate the risk of thromboembolic events because clinical manifestations of periprocedural thrombus formation, such as embolic stroke or valve thrombosis, may be delayed for several weeks after warfarin interruption.23 However, it is likely that most clinically overt bleeding episodes attributed to periprocedural bridging anticoagulation would be captured, as they are most likely to occur during the first week after a procedure.33,44-46,56 Second, we acknowledge that our classification of patients according to the procedure-related bleeding risk, which determined whether postprocedural dalteparin was given, was subjective and has not been validated. Such a classification of patients was deemed necessary to minimize postprocedural bleeding and was supported by studies that have identified high-bleeding-risk procedures.36-43 There are no clinical prediction rules, to our knowledge, that stratify patients according to the risk of postprocedural bleeding.57 Third, because this patient registry did not include a control group, we cannot comment on the efficacy or safety of our periprocedural anticoagulation regimen compared with other strategies that might involve intravenous heparin,11,12 or warfarin interruption without bridging anticoagulation.8,9 Finally, because this registry was undertaken at a single center that does not have a major focus on pulmonary or renal disease, there was underrepresentation of patients undergoing respiratory or nephrologic procedures. Consequently, our findings may not be generalizable to such patients, especially those with significant renal insufficiency in whom therapeutic-dose LMWH should be used with caution be- cause of the potential for bioaccumulation.58 Additional studies are needed to confirm our findings in patients who have a longer duration of postprocedural follow-up and to identify high-bleeding-risk procedures in which postprocedural anticoagulation should be used with caution. Furthermore, there is a need for clinical trials to compare our periprocedural anticoagulation regimen with less intense strategies (eg, low-dose LMWH) or alternative oral anticoagulants such as direct thrombin inhibitors, which, because of a short half-life, may obviate the need for bridging therapy with LMWH.59 To summarize, in patients at increased risk for arterial thromboembolism who require temporary interruption of warfarin therapy, a standardized periprocedural anticoagulant regimen with LMWH is associated with a low risk of thromboembolic and major bleeding complications in the immediate periprocedural period, and is feasible for outpatient administration. Additional studies are needed to confirm these findings and investigate alternative periprocedural anticoagulation strategies. Accepted for publication August 25, 2003. There was no funding for this study. Dr Douketis is a recipient of a Research Scholarship from the Heart and Stroke Foundation of Canada, Ottawa, Canada. We express sincere gratitude to Nicole Archer for establishing the patient registry database and to Nicole Zytaryk and Lauren Griffith for statistical assistance. We also thank Mark Crowther, MD, MSc, and Clive Kearon, MD, PhD, for their helpful reviews of the manuscript. Corresponding author: James D. Douketis, MD, FRCPC, Department of Medicine, St Joseph’s Hospital, Room F-541, 50 Charlton Ave E, Hamilton, Ontario, Canada L8N 4A6 (e-mail: [email protected]). (REPRINTED) ARCH INTERN MED/ VOL 164, JUNE 28, 2004 1325 REFERENCES 1. Dunn AS, Turpie AG. Perioperative management of patients receiving oral anticoagulants: a systematic review. Arch Intern Med. 2003;163:901-908. WWW.ARCHINTERNMED.COM ©2004 American Medical Association. All rights reserved. 2. Spandorfer J. The management of anticoagulation before and after procedures. Med Clin North Am. 2001;85:1109-1116. 3. Heit JA. Perioperative management of the chronically anticoagulated patient. J Thromb Thrombolysis. 2001;12:81-87. 4. Douketis JD. Perioperative anticoagulation management in patients who are receiving oral anticoagulant therapy: a practical guide for clinicians. Thromb Res. 2002;108:3-13. 5. Tiede DJ, Nishimura RA, Gastineau DA, Mullany CJ, Orszulak TA, Schaff HZ. Modern management of prosthetic valve anticoagulation. Mayo Clin Proc. 1998;73: 665-680. 6. Martinelli J, Jiminez A, Rabago G, Artiz V, Fraile J, Farre J. Mechanical cardiac valve thrombosis: is thrombectomy justified? Circulation. 1991;84(suppl 3):70-75. 7. Longstreth WT Jr, Bernick C, Fitzpatrick A, et al. Frequency and predictors of stroke death in 5,888 participants in the Cardiovascular Health Study. Neurology. 2001;56:368-375. 8. Kearon C, Hirsh J. Management of anticoagulation before and after elective surgery. N Engl J Med. 1997;336:1506-1511. 9. Hewitt RL, Chun KL, Flint LM. Current clinical concepts in perioperative anticoagulation. Am Surg. 1999;65:270-273. 10. Madura JA, Rookstool M, Wease G. The management of patients on chronic coumadin therapy undergoing subsequent surgical procedures. Am Surg. 1994;60: 542-547. 11. Busuttil B, Fabri BM. The management of anticoagulation in patients with prosthetic heart valves undergoing non-cardiac operations. Postgrad Med J. 1995; 71:390-392. 12. Bryan AJ, Butchart EG. Prosthetic heart valves and anticoagulant management during non-cardiac surgery. Br J Surg. 1995;82:577-578. 13. Bonow RO, Carabello B, de Leon AC, et al. ACC/AHA guidelines for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on Management of Patients With Valvular Heart Disease). J Am Coll Cardiol. 1998;32:1486-1588. 14. Ansell J, Hirsh J, Dalen J, et al. Managing oral anticoagulant therapy. Chest. 2001; 119(1, suppl):22S-38S. 15. Douketis JD, Crowther MA, Cherian S, Kearon C. Physician preferences for perioperative anticoagulation in patients with a mechanical heart valve who are undergoing elective non-cardiac surgery. Chest. 1999;116:1240-1246. 16. Douketis JD, Crowther MA, Cherian SS. Perioperative anticoagulation in patients with chronic atrial fibrillation undergoing elective surgery: results of a physician survey. Can J Cardiol. 2000;16:326-330. 17. Hirsh J, Warkentin TE, Shaughnessy SG, et al. Heparin and low-molecularweight heparin: mechanism of action, pharmacokinetics, dosing, monitoring, efficacy, and safety. Chest. 2001;119(1, suppl):64S-94S. 18. Goldstein JL, Larson LR, Yamashita BD, Fain JM, Schumock GT. Low molecular weight heparin versus unfractionated heparin in the colonoscopy periprocedure period: a cost modeling study. Am J Gastroenterol. 2001;96:23602366. 19. Katholi RE, Nolan SP, McGuire LB. Subsequent noncardiac operations and the risk of thromboembolism or hemorrhage. Am Heart J. 1976;92:162-167. 20. Katholi RE, Nolan SP, McGuire LB. The management of anticoagulation during noncardiac operations in patients with prosthetic heart valves. Am Heart J. 1978; 96:163-165. 21. Tinker JH, Tarhan S. Discontinuing anticoagulation therapy in surgical patients with cardiac valve prostheses. JAMA. 1978;239:738-739. 22. Darmon D, Enriquez-Sarano M, Acar J. Cardiac complications after subsequent non-cardiac operations in patients with non-biological prosthetic heart valves [abstract]. Eur J Cardiol. 1983;4(suppl 1):30. 23. Carrel TP, Klingenmann W, Mohacsi PJ, Berdat P, Althaus U. Perioperative bleeding and thromboembolic risk during non-cardiac surgery in patients with mechanical prosthetic heart valves: an institutional review. J Heart Valve Dis. 1999; 8:392-398. 24. Manley HJ, Smith JA, Garris RE. Subcutaneous enoxaparin for outpatient anticoagulation therapy in a patient with an aortic valve replacement. Pharmacotherapy. 1998;18:408-412. 25. Spandorfer JM, Lynch S, Weitz HH, Fertel S, Merli GJ. Use of enoxaparin for the chronically anticoagulated patient before and after procedures. Am J Cardiol. 1999; 84:478-480. 26. Tinmouth AH, Morrow BH, Cruickshank MK, Moore PM, Kovacs MJ. Dalteparin as periprocedure anticoagulation for patients on warfarin and at high risk of thrombosis. Ann Pharmacother. 2001;35:669-674. 27. Wahl MJ. Dental surgery in anticoagulated patients. Arch Intern Med. 1998;158: 1610-1616. 28. Horlocker TT, Wedel DJ, Benzon H, et al. Regional anesthesia in the anticoagulated patient: defining the risk (the Second ASRA Consensus Conference on Neuraxial Anesthesia and Anticoagulation). Reg Anesth Pain Med. 2003;28:172-197. 29. Anderson DR, Levine MN, Gent M, et al. Validation of new criteria for the evaluation of bleeding during antithrombotic therapy [abstract]. Thromb Haemost. 1993; 69:650. 30. Kasper W, Konstantinides S, Geibel A, et al. Management strategies and determinants of outcome in acute major pulmonary embolism: results of a multicenter registry. J Am Coll Cardiol. 1997;30:1165-1171. 31. Goldhaber SZ, Visani L, De Rosa M. Acute pulmonary embolism: clinical out- 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. (REPRINTED) ARCH INTERN MED/ VOL 164, JUNE 28, 2004 1326 comes in the International Cooperative Pulmonary Embolism Registry (ICOPER). Lancet. 1999;353:1386-1389. Mangano DT, for the Multicenter Study of Perioperative Bypass Surgery. Aspirin and mortality from coronary artery bypass surgery. N Engl J Med. 2002;347: 1309-1317. Pulmonary Embolism Prevention (PEP) Trial Collaborative Group. Prevention of pulmonary embolism and deep vein thrombosis with low dose aspirin: Pulmonary Embolism Prevention (PEP) Trial. Lancet. 2000;355:1295-1302. Collet JP, Montalescat G, Lison L, et al. Percutaneous coronary interventions after subcutaneous enoxaparin pretreatment in patients with unstable angina pectoris. Circulation. 2001;103:658-663. Dafnis G, Ekbom A, Pahlman L, Blomqvist P. Complications of diagnostic and therapeutic colonoscopy within a defined population in Sweden. Gastrointest Endosc. 2001;54:302-309. Milas BL, Jobes DR, Gorman RC. Management of bleeding and coagulopathy after heart surgery. Semin Thorac Cardiovasc Surg. 2000;12:326-336. Hingiorami A, Gramse C, Ascher E. Anticoagulation with enoxaparin versus intravenous unfractionated heparin in postoperative vascular surgery patients. J Vasc Surg. 2002;36:341-345. Lazio BE, Simard JM. Anticoagulation in neurosurgical patients. Neurosurgery. 1999;45:838-847. Cohen AT, Wagner MB, Mohamed MS. Risk factors for bleeding in major abdominal surgery using heparin thromboprophylaxis. Am J Surg. 1997;174:1-5. Dickinson LD, Miller LD, Patel CP, Gupta SK. Enoxaparin increases the incidence of postoperative intracranial hemorrhage when initiated preoperatively for deep vein thrombosis prophylaxis in patients with brain tumors. Neurosurgery. 1998;43:1074-1081. Chakravarti A, MacDermott S. Transurethral resection of the prostate in the anticoagulated patient. Br J Urol. 1998;81:520-522. Hergesell O, Felten H, Andrassy K, Kuhn K, Ritz E. Safety of ultrasound-guided percutaneous renal biopsy: retrospective analysis of 1090 consecutive cases. Nephrol Dial Transplant. 1998;13:975-977. Dillioglugil O, Liebman BD, Leibman NS, Kattan MW, Rosas AL, Scardino PT. Risk factors for complications and morbidity after radical prostatectomy. J Urol. 1997;157:1760-1767. Turpie AGG, Bauer KA, Eriksson BI, Lassen MR, for the Steering Committees of the Pentasaccharide Orthopedic Prophylaxis Studies. Fondaparinux vs enoxaparin for the prevention of venous thromboembolism in major orthopedic surgery: a meta-analysis of 4 randomized double-blind studies. Arch Intern Med. 2002;162:1833-1840. Mismetti P, Laporte S, Darmon J-Y, Buchmuller A, Decousus H. Meta-analysis of low molecular weight heparin in the prevention of venous thromboembolism in general surgery. Br J Surg. 2001;88:913-930. Koch A, Ziehler S, Breitschwerdt H, Victor N. Low molecular weight heparin and unfractionated heparin in thrombosis prophylaxis: meta-analysis based on original patient data. Thromb Res. 2001;102:295-309. Strebel N, Prins M, Agnelli G, Buller HR. Preoperative or postoperative start of prophylaxis for venous thromboembolism with low-molecular-weight heparin in elective hip surgery? Arch Intern Med. 2002;162:1451-1456. Michaud GF, Pelosi F Jr, Noble MD, Knight BP, Morady F, Strickberger SA. A randomized trial comparing heparin initiation 6h or 24h after pacemaker or defibrillator implantation. J Am Coll Cardiol. 2000;35:1915-1918. van Os EC, Kamath PS, Gostout CJ, Heit JA. Gastroenterological procedures among patients with disorders of hemostasis: evaluation and management recommendations. Gastrointest Endosc. 1999;50:536-543. Sandset PM, Abildgaard U. Perioperative management of oral anticoagulant therapy. Thromb Res. 2002;108:1-2. Genewein U, Haeberli A, Straub PW, Beer JH. Rebound after cessation of oral anticoagulant therapy: the biochemical evidence. Br J Haematol. 1996;92:479-485. Grip L, Blomback M, Schulman S. Hypercoagulable state and thromboembolism following warfarin withdrawal in post-myocardial infarction patients. Eur Heart J. 1991;12:1225-1233. Valles J, Aznar J, Santos T, Villa P, Fernandez A. Platelet function in patients with chronic coronary heart disease on long-term anticoagulant therapy: effect of anticoagulant therapy. Haemostasis. 1993;23:212-218. Freyburger G, Janvier G, Dief S, Biousseau MR. Fibrinolytic and hemostatic alterations during and after aortic graft surgery: implications for postoperative management. Anesth Analg. 1993;76:504-512. D’Angelo A, Kluft C, Verheijen JH, Rijken DC, Mozzi E, Mannucci PM. Fibrinolytic shutdown after surgery: impairment of the balance between tissue-type plasminogen activator and its specific inhibitor. Eur J Clin Invest. 1985;15:308-312. Ferreira I, Dos L, Tornos P, Nicolau I, Permanyer-Miralda G, Soler-Soler J. Experience with enoxaparin in patients with mechanical heart valves who must withhold acenocoumarol. Heart. 2003;89:527-530. Levine MN, Raskob GE, Landefeld SC, Kearon C. Hemorrhagic complications of anticoagulant treatment. Chest. 2001;119(1, suppl):108S-121S. Nagge J, Crowther M, Hirsh J. Is impaired renal function a contraindication to the use of low-molecular-weight heparin? Arch Intern Med. 2002;162:26052609. Gustafsson D, Elg M. The pharmacodynamics and pharmacokinetics of the oral direct thrombin inhibitor ximelegatran and its active metabolite melagatran: a mini-review. Thromb Res. 2003;109(suppl 1):S9-S15. WWW.ARCHINTERNMED.COM ©2004 American Medical Association. All rights reserved.