Prospect
Prospect
Prospect
ORIGINAL ARTICLE
To cite this article: Dunn AS, Spyropoulos AC, Turpie AGG. Bridging therapy in patients on long-term oral anticoagulants who require surgery: the
Prospective Peri-operative Enoxaparin Cohort Trial (PROSPECT). J Thromb Haemost 2007; 5: 2211–8.
peri-operative strategy using once-daily therapeutic-dose Patients were screened within 28 days of the surgery or
enoxaparin administered primarily at home, and the effect, if procedure, including a medical history, physical examination,
any, of the extensiveness of the procedure on the risk of and clinical laboratory testing. Per protocol, the intention-to-
bleeding during bridging therapy. We also sought to examine treat (ITT) study population was defined as all patients who
the incidence of arterial thromboembolic events for patients received at least one dose of enoxaparin. All analyses were
with atrial fibrillation and the incidence of venous thrombo- performed on this population. Procedures were a priori
embolic events for patients with a history of DVT. categorized as an invasive procedure, minor surgery, or major
surgery. Major surgery was defined as any operation that had
an expected duration ‡ 1 h. Classification of a procedure as an
Methods
invasive procedure or minor surgery was at the discretion of the
site investigator.
Study design
This was a prospective, multicenter, cohort study evaluating
Sample size calculation
the efficacy and safety of therapeutic-dose enoxaparin, admin-
istered primarily at home, in patients on long-term warfarin for The sample size was determined to achieve a precise estimate of
whom bridging therapy with heparin was planned for an the incidence of peri-operative major bleeding. As the bleeding
invasive procedure or surgery. The study was conducted in 24 rate is unknown, the estimated incidence of major bleeding was
study sites in North America between January 2002 and based on limited data from prior studies [4,10]. For an expected
August 2003. The primary aim was to determine the incidence frequency of major bleeding of 2%, a sample size of 300 would
of peri-operative major bleeding during administration of provide a 95% CI of ± 1.6%.
therapeutic-dose bridging anticoagulation, and whether the
bleeding rate is affected by the extensiveness of the procedure.
Peri-operative management of anticoagulation
The study protocol was reviewed and approved by the
independent Ethics Committee or Institutional Review Board The peri-operative management of anticoagulation is outlined
for each center. Written formal consent was obtained from all in Fig. 1. Warfarin was discontinued 5 days before surgery
patients enrolled in the study. (that is, the last dose was given 6 days prior to the day of
Study sample
Day-5 prior to procedure
Outpatients between the ages of 18 and 85 years, with a body Warfarin discontinuation
weight ‡ 45 kg and £ 120 kg, who required chronic warfarin
for atrial fibrillation or prior DVT, who required interruption Day-3 to Day-1 prior to procedure.
of warfarin for major or minor surgery or an invasive Start enoxaparin at home in the morning (9 am ± 2 h)
procedure, and for whom peri-operative bridging therapy with 1.5 mg/kg s.c. qd
therapeutic-dose anticoagulation was planned by their primary
physician were eligible for inclusion.
Day 0 - day of procedure
Patients were excluded if they had mechanical heart valves, Enoxaparin discontinuation
ischemic stroke in the month prior to enrollment, or any
previous hemorrhagic stroke; active bleeding, recent gastroin-
testinal bleeding, or a congenital or acquired bleeding disorder;
history of intraocular bleeding or proliferative diabetic reti- Day 0 - evening of procedure
Restart warfarin if there are no contra-indications
nopathy; anemia considered significant by the investigator or
hemoglobin < 11.0 g dL)1 if male, or < 10.0 g dL)1 if
female; platelet count < 100 000 per mm3; intracranial tumor,
aneurysm, or arteriovenous malformation; severe renal disease 12-24 h post-procedure
(calculated creatinine clearance £ 30 mL min)1 or serum Restart enoxaparin 1.5 mg/kg s.c. qd provided
there is no active surgical bleeding
creatinine > 2.0 mg dL)1); severe hepatic impairment; uncon-
trolled hypertension; cardiogenic shock; active endocarditis;
carotid artery surgery, intraocular surgery, or neurosurgery;
planned spinal/epidural anesthesia or puncture during the peri- Enoxaparin continuation until INR is therapeutic
operative period; life-expectancy < 3 months; or any factor
that would impede compliance with treatment or home
monitoring. Patients with known sensitivity to pork products,
Patient follow up at 28 days after therapeutic INR
murine proteins, unfractionated heparin or low molecular
weight heparin or any of its constituents, or history of heparin- Fig. 1. Peri-operative bridging protocol with low-molecular-weight
induced thrombocytopenia, were also excluded. heparin enoxaparin. INR, International Normalized Ratio.
surgery). Starting 3 days prior to the procedure, enoxaparin arterial thromboembolism. Venous thromboembolic events
was administered s.c. each morning (9 AM ± 2 h) at home at a were defined as acute symptomatic DVT (documented by
dose of 1.5 mg kg)1. The final pre-operative dose was admin- compression ultrasound or venogram) or acute symptomatic
istered on the morning prior to the procedure. On the day prior pulmonary embolism (documented by CT scan, pulmonary
to, or on the morning of, the procedure, the prothrombin time angiogram, or ventilation/perfusion scan, and if necessary
and International Normalized Ratio (INR) were measured. substantiated by abnormal ultrasound or venogram). Arterial
Vitamin K (1–2.5 mg orally) was given if the INR was ‡ 1.8. events that occurred in patients without history of atrial
Warfarin was restarted on the evening of the procedure or on fibrillation (i.e. enrolled because of history of DVT) and venous
the following day, provided that adequate hemostasis was events that occurred in patients without history of DVT (i.e.
achieved. The dose of warfarin was the same as the patientÕs enrolled because of history of atrial fibrillation) were recorded
usual daily dose prior to hospitalization. Subcutaneous enox- but not considered part of the primary efficacy endpoint.
aparin was reinitiated at a dose of 1.5 mg kg)1 once-daily 12–
24 h after the procedure. The first post-operative dose of
Results
enoxaparin could be delayed if hemostasis was inadequate,
based on the assessment of the surgeon/procedurist and site In total, 283 subjects were enrolled between January 2002 and
investigator. Treatment with enoxaparin was continued until August 2003. This was 17 patients short of the projected sample
the INR was within the target range. size, because of slower than expected enrollment. Of the
Hemoglobin and platelet levels were measured every 1– enrolled patients, 23 did not receive study medication (Fig. 2).
2 days while patients were on enoxaparin. The INR was In total, 260 patients received at least one dose of enoxaparin,
measured every 1–2 days during the first week after the and comprised the ITT study population. Efficacy and safety
procedure. The follow-up period extended from 24 h after a analyses were based on the ITT population. Ten patients did
therapeutic INR was achieved to 1 month (28 ± 4 days) not undergo the planned procedure (three undergoing an
thereafter. invasive procedure, four for minor surgery, and three for major
surgery). Enoxaparin was discontinued prematurely for 48
patients (18.5%), most commonly for the occurrence of an
Outcomes
adverse event (Fig. 2). The most common adverse event
The primary safety outcome was the incidence of major leading to discontinuation of enoxaparin was minor bleeding,
bleeding while on enoxaparin or in the 24 h following cessation which led to 19 withdrawals for the following reasons:
of enoxaparin treatment. Major hemorrhage was defined as hematoma (n = 6); macroscopic hematuria (n = 5); ecchy-
overt bleeding leading to a ‡ 3 g dL)1 drop in hemoglobin mosis (n = 3); and other reasons (n = 5). Enoxaparin
(transfusion of 1 unit of red blood cells or whole blood was discontinuation was due to major bleeding for four patients
considered to be equivalent to a 1 g dL)1 in decrease in (two episodes of incision site bleeding, one abdominal hema-
hemoglobin), transfusion of ‡ 2 units of packed red blood cells toma, and one gastrointestinal hemorrhage).
or whole blood, or any bleeding that was intracranial, Baseline characteristics of the study population are shown in
retroperitoneal, or intraocular, that required surgical interven- Table 1. The mean age was 68 years. Approximately two-
tion, or that resulted in death. All suspected major bleeds were thirds of the population received long-term OAC therapy for
adjudicated by the study steering committee. A planned atrial fibrillation, and one-third for prior DVT. An assessment
subgroup analysis was performed to determine the incidence of the overall stroke risk for the patients with a history of atrial
of major bleeding by extensiveness of procedure (invasive fibrillation using the CHADS risk index is shown in Table 2
procedures, minor surgery, or major surgery). The secondary [11]. Patients enrolled in the study had a distribution of
safety outcome was the rate of minor bleeding while on CHADS scores that was similar to that noted in the National
enoxaparin, or within 24 h of discontinuation. All bleeding Registry of Atrial Fibrillation Patients (for example, 33% and
events that did not meet the criteria for major bleeding were 36% had a CHADS score of 3 or higher, respectively) [11]. One
classified as minor. hundred and forty-eight patients (56.9%) required an invasive
The primary efficacy outcomes were the incidence of arterial procedure, 72 (27.7%) minor surgery, and 40 (15.4%) required
thromboembolic events for patients with atrial fibrillation and major surgery (Table 3). No patients required pre-operative
the incidence of venous thromboembolic events for patients vitamin K administration. Enoxaparin was administered at
with a history of DVT. Thromboembolic events were included home for 85.8% of doses. Of the ITT population, 245 subjects
if they occurred at any time during the study period (from the (94.2%) received postprocedure enoxaparin for a mean of
cessation of warfarin therapy 5 days prior to surgery until 6.8 days. Enoxaparin was started at a similar time postoper-
28 days after the INR was within the therapeutic range). atively in the three groups (1.1 ± 0.64 days, 1.10 ± 0.46 days,
Arterial thromboembolic events were defined as ischemic and 1.17 ± 0.78 days, in the invasive procedure, minor
stroke [documented by computed tomography (CT) scan or surgery and major surgery groups, respectively).
magnetic resonance imaging], transient ischemic attacks (TIAs) The overall incidence of major bleeding while patients
(documented by the treating physicians and including lack of were on enoxaparin plus the following 24 h was 3.5% (95% CI:
evidence of an acute stroke on brain imaging), or peripheral 1.6–6.5) (Table 4). The bleeding risk varied markedly by
Fig. 2. Flowchart of patient enrollment and procedures. *These subjects were still considered part of the intention-to-treat population.
Age, mean ± SD (years) 68.1 ± 11.1 Invasive procedure (%) 148 (56.9)
Gender Cardiovascular 77
Male (%) 147 (56.5) Cardiac catheterization 57
Female (%) 113 (43.5) Pacemaker implantation 7
Weight, mean ± SD (kg) 86.0 ± 18.8 Implantable cardiac defibrillator 3
Systolic blood pressure, 130.5 ± 19.3 Radiofrequency ablation 3
mean ± SD (mmHg) Other 7
Diastolic blood pressure, 74.2 ± 9.7 Gastrointestinal 46
mean ± SD (mmHg) General 7
Heart rate, mean ± SD 71.9 ± 12.8 Urologic 7
(beats per minute) Dental 4
Race Orthopedic 2
Caucasian (%) 232 (89.2) Other 4
Black (%) 15 (5.8) Skin 1
Asian (%) 1 (0.4) Minor surgery (%) 72 (27.7)
Hispanic (%) 10 (3.8) Dental 22
American Indian (%) 1 (0.4) Cardiovascular 11
Other (%) 1 (0.4) Gastrointestinal 8
Reason for anticoagulation* Orthopedic 7
Atrial fibrillation (%) 176 (67.7) Ear, nose, and throat 5
History of deep vein thrombosis (%) 96 (36.9) General 4
Skin 4
*Sums to greater than 100% because the categories are not mutually Breast 4
exclusive. Urologic 5
Ocular 2
Table 2 Risk of stroke for patients with atrial fibrillation using the Major surgery (%) 40 (15.4)
CHADS risk index* Orthopedic 19
CHADS score n (%) Cardiovascular 9
General 6
0 21 (11.9) Gynecologic 3
1 57 (32.4) Gastrointestinal 2
2 40 (22.7) Urologic 1
3 35 (19.9)
4 20 (11.4)
5 3 (1.7) extensiveness of procedure: the incidence of major bleeding for
6 0 invasive procedures, minor surgery, and major surgery was
*Score calculated as 1 point for each of congestive heart failure,
0.7% (95% CI: 0.02–3.7), 0% (95% CI: 0–5.0), and 20.0%
hypertension, age > 75 years, or diabetes; and 2 points for a prior (95% CI: 9.1–35.7), respectively. Most major bleeding events
history of stroke or transient ischemic attack [11]. were overt surgical site bleeding associated with a ‡ 3 g dL)1
decrease in hemoglobin or requiring transfusion of ‡ 2 U of in platelet count below 100 000 per mm3 while on enoxaparin
packed red blood cells. None of the hemorrhages was (range 64 000–99 000). No patients satisfied the criteria for
intracranial, retroperitoneal, intraocular, or fatal, or required heparin-induced thrombocytopenia.
surgical intervention. None of the patients who experienced Thromboembolic complications occurred in five patients
major bleeding had received a dose of enoxaparin on the day of (1.9%, 95% CI: 0.6–4.4) (Table 5). Three events took place
the surgery. Among those undergoing major surgery, five out during the peri-operative period, and two occurred during
of the eight major bleeding events occurred following total hip follow-up. Four arterial events occurred in 176 patients
or knee replacement. Examination of major bleeding rates for (2.3%, 95% CI: 0.6–5.7) with atrial fibrillation. There were
the entire study period, including the 28-day follow-up period, two TIAs, there were no strokes, and two patients experi-
revealed similar results (1.4%, 0%, and 27.5%, for invasive enced peripheral arterial thromboembolic events. One pul-
procedures, minor surgery, and major surgery, respectively). monary embolism (1.0%, 95% CI: 0.03–5.7) was noted in 96
Minor bleeding occurred in 108 patients (41.5%, 95% CI: patients with a history of DVT. None of the thromboembolic
35.7–47.6). In contrast to major bleeding, there was no increase events was fatal. One of the five thromboembolic events
in the incidence of minor bleeding based on the extensiveness of occurred after bleeding that led to withdrawal of enoxaparin
the procedure. The incidence of minor bleeding for patients and warfarin. Two other thromboembolic events were noted
undergoing invasive procedures, minor surgery and major that were not part of the predefined primary efficacy
surgery was 44.6%, 47.2%, and 20.0%, respectively. The most outcome; a DVT in a patient with atrial fibrillation who
common minor bleeding events were ecchymosis (63.9%), had no history of DVT, and a peripheral arterial thrombo-
hematoma (23.1%), macroscopic hematuria (7.4%), and embolism in a patient with a history of DVT who did not
epistaxis (7.4%). Six patients (2.3%) experienced a decrease have atrial fibrillation.
Atrial fibrillation TIA 55-year-old male with a history of atrial TIA occurred immediately following major
fibrillation and stroke surgery for cardiac ablation
Atrial fibrillation TIA 67-year-old male with atrial fibrillation TIA occurred immediately following
cardiac catheterization
Atrial fibrillation Leg ischemia 76-year-old male with a history of atrial Required aortoiliofemoral thrombectomy and
fibrillation, hypertension, peripheral arterial femoral popliteal bypass surgery after major
disease, and DVT surgical repair of an endovascular abdominal
aortic aneurysm
Atrial fibrillation Peripheral arterial 76-year-old male with a history of atrial The patient experienced major bleeding 7 days
thromboembolism fibrillation, hypertension, and mild after total knee arthroplasty, at which time
renal insufficiency warfarin and enoxaparin were discontinued.
Six days after the onset of the major bleed, the
patient developed a peripheral arterial
thromboembolism
DVT Pulmonary embolism 72-year-old male with a history of DVT, Pulmonary embolism occurred 9 days after the
thyroid cancer, hypertension, and TIA last dose of enoxaparin after undergoing
hemithyroidectomy
DVT, deep vein thrombosis; OAC, oral anticoagulant; TIA, transient ischemic attack.
None of the 23 patients who were enrolled but did not (including studies of venous thromboembolism prophylaxis)
receive any doses of enoxaparin (and thus were not part of the have generally found lower rates than found in our study while
ITT population) had a major bleeding or thromboembolic using similar definitions for major bleeding. Finally, the
event. relatively small number of patients undergoing major surgery
could have allowed chance to be a factor in the observed event
rate.
Discussion
This study represents the largest prospective clinical trial of
An estimate of the risk of postoperative major bleeding is bridging therapy in which a uniform peri-operative bridging
essential for clinicians and patients to make an informed strategy was used for all patients. Performance at 24 academic
decision about the risks and benefits of administering full-dose and community hospitals across North America facilitates
parenteral anticoagulation peri-operatively while OAC is broad generalizability of the results to other settings. Most
subtherapeutic. Although reviews and cohort studies have notably, the prospective design, standardized strategy for all
suggested that bridging anticoagulation may cause a greater procedure types and a priori definition of procedures as
increase in major bleeding after major surgery relative to less invasive procedures, minor surgery and major surgery allowed
extensive procedures, and have accounted for this perceived examination of the safety of the bridging strategy for proce-
increased risk by recommending or administering decreased dures of varying extensiveness. Other studies have not been
anticoagulant doses after major surgery [5,6,12], there has been able to provide clinically meaningful estimates of the incidence
little or no direct examination of this issue. Our study of major bleeding on treatment-dose bridging anticoagulation,
confirmed that the bleeding risk is high when bridging therapy because of combining bleeding rates for different types of
is administered peri-operatively to patients undergoing major procedures, and utilization of different strategies for different
surgery. Major bleeding occurred in eight of 40 patients (20%) types of procedure [4–6].
undergoing major surgery, as compared with one major Low molecular weight heparin as bridging therapy has been
bleeding event in 220 patients (0.5%) undergoing invasive evaluated in several prior studies [4–6,8,15]. These studies have
procedures or minor surgery. This indicates that other strat- used a variety of designs and found a wide range of results,
egies need to be evaluated to determine the optimal manage- including major bleeding rates ranging from 0.7% to 15%.
ment of high-risk patients with atrial fibrillation or prior Douketis et al. [6] examined 650 consecutive patients with
venous thromboembolism undergoing major surgery, particu- mechanical heart valves, atrial fibrillation or embolic stroke
larly in the postoperative period. Our finding that the incidence who received peri-operative dalteparin, and found a low
of major bleeding is low for patients receiving the therapeutic incidence of major bleeding and thromboembolic events (0.7%
once-daily dose of enoxaparin undergoing invasive procedures and 0.4%, respectively). Dalteparin was not administered
or minor surgery, encompassing 84.6% of the study popula- post-operatively for patients who underwent a Ôhigh bleeding
tion, suggests that this regimen can be used safely for most riskÕ procedure. In contrast, a retrospective analysis of mem-
patients undergoing procedures. These categories comprised a bers of a managed care organization undergoing bridging
large and diverse group of procedures, including inguinal therapy for primarily major surgery found a major bleeding
hernia repair, breast lumpectomy, laparoscopic gallbladder rate in patients bridged with enoxaparin of 15%, similar to the
resection, dental extractions, pacemaker insertion, and tran- rate in our study for patients undergoing major surgery [8]. A
surethral resection of bladder tumor. Other studies have also large, prospective, multicenter registry of patients receiving
found that most patients being considered for bridging therapy chronic OACs requiring peri-operative bridging for an elective
will undergo invasive procedures or minor surgery rather than procedure or surgery reported rates of thromboembolism of
major surgery, supporting the suggestion that the regimen 2.4% and 0.9%, and rates of major bleeding of 5.5% and
studied is applicable to most patients who require bridging 3.3%, for unfractionated heparin and low molecular weight
[4–6]. heparin, respectively [15]. A prospective trial of patients with
Several factors may explain the high incidence of major atrial fibrillation or mechanical heart valves assessed the use of
bleeding for patients undergoing major surgery. First, it is peri-operative bridging therapy with therapeutic-dose daltep-
possible that the last dose administered pre-operatively could arin for patients not at high bleeding risk, and prophylaxis-dose
have produced a residual anticoagulant effect at the time of dalteparin for patients felt to be at high risk for bleeding. This
surgery. Such a residual effect of low molecular weight heparin strategy was associated with major bleeding and thromboem-
has been demonstrated in 16% of bridging cases just before the bolic event rates of 6.7% and 3.6%, respectively [5]. Overall,
procedure, and can be predicted by therapeutic dose and these studies provide little guidance for clinicians, because of
increasing age [13,14]. Post-operatively, the once-daily dose of the regimens being based on perceived risk of bleeding and a
enoxaparin could have produced a high peak level at a point wide range of event rates.
where healing at the incision site was at a vulnerable stage. It is The primary limitation of the study is the lack of a control
plausible that the event rate may have been elevated by vigilant group that did not receive peri-operative enoxaparin. Thus, we
examination and reporting of ÔnormalÕ blood loss during major cannot determine the efficacy of enoxaparin in decreasing the
surgery. This seems unlikely, however, as other studies incidence of venous or arterial thromboembolism. In addition,
examining the incidence of major bleeding in major surgery patients with mechanical heart valves were excluded, as at the
time when the study was being designed, case reports raised
Acknowledgements
concerns over the safety of low molecular weight heparin for
patients with mechanical heart valves [16,17], and it was felt The authors thank M. Chen (Sanofi-Aventis U.S. Inc.) for help
that further information was needed before including patients with the statistical analyses.
with mechanical heart valves in a large-scale study using low
molecular weight heparin. An additional limitation is the lack
Disclosure of Conflict of Interests
of power to make a precise estimate of the incidence of
thromboembolic events, most notably the stroke rate, during Funding for the study and editorial support for the publication
the bridging period. We did not gather systematic data on were provided by sanofi-aventis US Inc.
patients who were bridged at the study sites who were not
enrolled in the study, and thus are unable to determine how
Appendix
representative the study population is of all patients being
considered for bridging. However, patients in our study had a The following investigators and centers participated in this trial:
wide distribution of CHADS scores (Table 2), with a similar A. Dunn, Mount Sinai School of Medicine, New York, NY; S.
distribution as noted in the original derivation of the CHADS K. Elliott, MediSphere Medical Research Center, Evansville,
scoring system (based on data from the National Registry of IN; G. Emlein, Sacramento Heart Research Department,
Atrial Fibrillation) [11]. In addition, patients had an average Sacramento, CA; P. Gainey, St Joseph Hospital, Savannah,
age, weight and distribution of procedures similar to that noted GA; D. Green, New York University School of Medicine, New
in the REGIMEN observational registry of patients under- York, NY; J. B. Groce III, Moses Cone Health System,
going bridging anticoagulation, also suggesting that our Greensboro, NC; M. Hollman, Florida Orthopedic Associates,
population was representative of patients undergoing bridging Orange City, FL; A. Jaffer, The Cleveland Clinic, Cleveland,
therapy and not skewed towards low risk for bleeding or OH; R. Lambert, Pulmonary and Research Associates,
thromboembolism [15]. Spokane, WA; J. R. LaSalle, Medical Arts Research Collabo-
Although patients undergoing dental procedures were rative, Excelsior Springs, MO; W. Wilson, Stucky Research
included to mimic clinical practise, where patients undergoing Center Parkview Hospital, Fort Wayne, IN; G. V. Naccarelli,
these procedures often receive bridging therapy, the inclusion Penn State University College of Medicine, Hershey, PA; R.
of dental procedures could contribute to low bleeding rates. OÕRourke, Audie L. Murphey Veterans Affairs Hospital, San
There were 26 patients who underwent dental procedures (22 Antonio, TX; N. Perlmutter, Overlake Internal Medicine
classified as minor surgery and four as invasive procedures), Associates, Belleview, WA; H. Punatar, Northern California
including single and multiple extractions, root canal treatment, Medical Associates, Santa Rosa, CA; M. Raja, Summit
and other procedures. Excluding these patients from the Research Solutions, Memphis, TN; R. Schumacher, Methodist
analysis does not change the major bleeding rates for each Research Institute, Indianapolis, IN; L. Siddoway, York
subgroup (0.7%, 0% and 20.0% for invasive procedures, Hospital, York, PA; J. Spandorfer, Thomas Jefferson Univer-
minor surgery, and major surgery, respectively), supporting the sity, Philadelphia, PA; A. C. Spyropoulos, Lovelace Health
finding of low bleeding rates for patients undergoing non- System, Albuquerque, NM; A. Graham Turpie, McMaster
major surgery. University, Hamilton, ON; P. Bove, William Beaumont
The appropriate post-operative management of patients at Hospital, Royal Oak, MI; D. Cragg, William Beaumont
high thromboembolic risk undergoing major surgery remains a Hospital, Troy, MI; S. K. Fitzmorris, Arrowhead Cardiology
clinical dilemma, as our study found a substantial risk of major Group, Corlton, CA; G. Garza, Rio Grande Valley Clinical
bleeding if bridging therapy is administered soon after surgery. Research Center, Edinburg, TX; P. Kakavas, Heart Care
Possible approaches in the post-operative period may be the Center of Illinois, Mokena, IL; S. Kaatz, Henry Ford
use of a twice-daily therapeutic dose of low molecular weight Health Systems, Detroit, MI; M. Cohen, Newark Beth Israel
heparin instead of the once-daily dose to avoid a potentially Medical Center, Newark, NJ; P. D. Garrett, Brooke
high peak concentration soon after surgery, initial administra- Army Medical Center, San Antonio, TX; E. van de Graaf,
tion of prophylactic doses with conversion to full dose when the Wright-Patterson Medical Center, Wright-Patterson Air Force
bleeding risk has felt to be diminished, and not administering Base, OH, USA.
post-operative bridging anticoagulation. The latter option may
be reasonable for patients who are felt not to be at high risk of
References
thromboembolism during a brief period when OAC treatment
is interrupted. 1 Stafford RS, Singer DE. Recent national patterns of warfarin use in
atrial fibrillation. Circulation 1998; 97: 1231–3.
In conclusion, our findings demonstrate that bridging
2 Levine MN, Raskob G, Landefeld S, Kearon C. Hemorrhagic com-
therapy with once-daily enoxaparin in an outpatient setting is plications of anticoagulant treatment. Chest 2001; 119 (Suppl. 1):
feasible and associated with a low incidence of bleeding events 108S–21S.
for patients undergoing minor surgery or invasive procedures. 3 Dunn AS, Turpie AG. Perioperative management of patients receiving
Further studies are needed to optimize the bridging regimen for oral anticoagulants: a systematic review. Arch Intern Med 2003; 163:
901–8.
patients undergoing major surgery.
4 Johnson J, Turpie AGG Temporary discontinuation of oral antico- 11 Gage BF, Waterman AD, Shannon W, Boechler M, Rich MW,
agulants: role of low-molecular-weight heparin (dalteparin). Thromb Radford MJ. Validation of clinical classification schemes for predicting
Haemost 1999; 82 (Suppl. Aug 1999): 62–3. stroke: results from the National Registry of Atrial Fibrillation. JAMA
5 Kovacs MJ, Kearon C, Rodger M, Anderson DR, Turpei AG, Bates 2001; 285: 2864–70.
SM, Desjardins L, Douketis J, Kahn SR, Sulymoss S, Wells PS. Single- 12 Jafri SM. Periprocedural thromboprophylaxis in patients receiving
arm study of bridging therapy with low-molecular-weight heparin for chronic anticoagulation therapy. Am Heart J 2004; 147: 3–15.
patients at risk of arterial embolism who require temporary inter- 13 Douketis JD, Woods K, Foster GA, Crowther MA. Bridging anti-
ruption of warfarin. Circulation 2004; 110: 1658–63. coagulation with low-molecular-weight heparin after interruption of
6 Douketis JD, Johnson JA, Turpie AG. Low-molecular-weight heparin warfarin therapy is associated with a residual anticoagulant effect prior
as bridging anticoagulation during interruption of warfarin: assess- to surgery. Thromb Haemost 2005; 94: 528–31.
ment of a standardized periprocedural anticoagulation regimen. Arch 14 OÕDonnell MJ, Kearon C, Johnson J, Robinson M, Zondag M, Turpie
Intern Med 2004; 164: 1319–26. I, Turpie AG. Preoperative anticoagulant activity after bridging low-
7 Douketis JD, Crowther MA, Cherian SS. Perioperative anti- molecular-weight heparin for temporary interruption of warfarin. Ann
coagulation in patients with chronic atrial fibrillation who are Intern Med 2007;146: 184–7.
undergoing elective surgery: results of a physician survey. Can J 15 Spyropoulos AC, Turpie AG, Dunn AS, Spandorfer J, Douketis J,
Cardiol 2000; 16: 326–30. Jacobson A, Frost FJ, REGIMEN Investigators. Clinical outcomes
8 Spyropoulos AC, Frost FJ, Hurley JS, Roberts M. Costs and clinical with unfractionated heparin or low-molecular-weight heparin as
outcomes associated with low-molecular-weight heparin vs unfrac- bridging therapy in patients on long-term oral anticoagulants: the
tionated heparin for perioperative bridging in patients receiving long- REGIMEN registry. J Thromb Haemost 2006; 4: 1246–52.
term oral anticoagulant therapy. Chest 2004; 125: 1642–50. 16 Idir M, Madonna F, Roudaut R. Collapse and massive pulmonary
9 Warkentin TE, Greinacher A. Heparin-induced thrombocytopenia: edema secondary to thrombosis of a mitral mechanical heart valve
recognition, treatment, and prevention: the Seventh ACCP Conference prosthesis during low-molecular weight heparin therapy. J Heart Valve
of Antithrombotic and Thrombolytic Therapy. Chest 2004; 126 Dis 1999; 8: 303–4.
(Suppl. 3): 311S–37S. 17 Lev-Ran O, Kramer A, Gurevitch J, Shapira I, Mohr R. Low-
10 Spandorfer JM, Lynch S, Weitz HH, Fertel S, Merli GJ. Use of molecular-weight heparin for prosthetic heart valves: treatment failure.
enoxaparin for the chronically anticoagulated patient before and after Ann Thorac Surg 2000; 69: 264–5.
procedures. Am J Cardiol 1999; 84: 478–80.