Portal de Periódicos da CAPES

Clinical outcomes with unfractionated heparin or low‐molecular‐weight heparin as bridging therapy in patients on long‐term oral anticoagulants: the REGIMEN registry

2006; Elsevier BV; Volume: 4; Issue: 6 Linguagem: Inglês

10.1111/j.1538-7836.2006.01908.x

1538-7933

Alex C. Spyropoulos, Alexander G.G. Turpie, Andrew Dunn, John Spandorfer, James D. Douketis, A. Jacobson, Floyd J. Frost,

Acute Myocardial Infarction Research

Journal of Thrombosis and HaemostasisVolume 4, Issue 6 p. 1246-1252 Free Access Clinical outcomes with unfractionated heparin or low-molecular-weight heparin as bridging therapy in patients on long-term oral anticoagulants: the REGIMEN registry1 A. C. SPYROPOULOS, A. C. SPYROPOULOS Lovelace Medical Center, Albuquerque, NM, USASearch for more papers by this authorA. G. G. TURPIE, A. G. G. TURPIE Hamilton Health Sciences General Hospital, Hamilton, ON, CanadaSearch for more papers by this authorA. S. DUNN, A. S. DUNN Mount Sinai Hospital, New York, NYSearch for more papers by this authorJ. SPANDORFER, J. SPANDORFER Thomas Jefferson University Hospital, Philadelphia, PA, USASearch for more papers by this authorJ. DOUKETIS, J. DOUKETIS St Joseph's Hospital, Hamilton, ON, CanadaSearch for more papers by this authorA. JACOBSON, A. JACOBSON Loma Linda VA Medical Center, Loma Linda, CASearch for more papers by this authorF. J. FROST, F. J. FROST Lovelace Respiratory Research Institute, Albuquerque, NM, USASearch for more papers by this authorTHE REGIMEN INVESTIGATORS, THE REGIMEN INVESTIGATORS Lovelace Medical Center, Albuquerque, NM, USASearch for more papers by this author A. C. SPYROPOULOS, A. C. SPYROPOULOS Lovelace Medical Center, Albuquerque, NM, USASearch for more papers by this authorA. G. G. TURPIE, A. G. G. TURPIE Hamilton Health Sciences General Hospital, Hamilton, ON, CanadaSearch for more papers by this authorA. S. DUNN, A. S. DUNN Mount Sinai Hospital, New York, NYSearch for more papers by this authorJ. SPANDORFER, J. SPANDORFER Thomas Jefferson University Hospital, Philadelphia, PA, USASearch for more papers by this authorJ. DOUKETIS, J. DOUKETIS St Joseph's Hospital, Hamilton, ON, CanadaSearch for more papers by this authorA. JACOBSON, A. JACOBSON Loma Linda VA Medical Center, Loma Linda, CASearch for more papers by this authorF. J. FROST, F. J. FROST Lovelace Respiratory Research Institute, Albuquerque, NM, USASearch for more papers by this authorTHE REGIMEN INVESTIGATORS, THE REGIMEN INVESTIGATORS Lovelace Medical Center, Albuquerque, NM, USASearch for more papers by this author First published: 17 May 2006 https://doi.org/10.1111/j.1538-7836.2006.01908.xCitations: 137 Alex C. Spyropoulos, Clinical Thrombosis Center, Lovelace Medical Center, 5400 Gibson Boulevard S.E., Albuquerque, NM 87108, USA.Tel.: +1 505 262 7874; fax: +1 505 262 7040; e-mail: alex.spyropoulos@lovelacesandia.com 1 This study has been presented in abstract form at the 46th American Society of Hematology meeting, San Diego, CA, 4–7 December, 2004. AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinked InRedditWechat Abstract Summary. Background: Patients who receive long-term oral anticoagulant (OAC) therapy often require interruption of OAC for an elective surgical or an invasive procedure. Heparin bridging therapy has been used in these situations, although the optimal method has not been established. No large prospective studies have compared unfractionated heparin (UFH) with low-molecular-weight heparin (LMWH) for the perioperative management of patients at risk of thromboembolism requiring temporary interruption of long-term OAC therapy. Patients/methods: This multicenter, observational, prospective registry conducted in North America enrolled 901 eligible patients on long-term OAC who required heparin bridging therapy for an elective surgical or invasive procedure. Practice patterns and clinical outcomes were compared between patients who received either UFH alone (n = 180) or LMWH alone (n = 721). Results: Overall, the majority of patients (74.5%) requiring heparin bridging therapy had arterial indications for OAC. LMWH, in mostly twice-daily treatment doses, represented approximately 80% of the study population. LMWH-bridged patients had significantly fewer arterial indications for OAC, a lower mean Charlson comorbidity score, and were less likely to undergo major or cardiothoracic surgery, receive intraprocedural anticoagulants or thrombolytics, or receive general anesthesia than UFH-bridged patients (all P < 0.05). The LMWH group had significantly more bridging therapy completed in an outpatient setting or with a < 24-h hospital stay vs. the UFH group (63.6% vs. 6.1%, P < 0.001). In the LMWH and UFH groups, similar rates of overall adverse events (16.2% vs. 17.1%, respectively, P = 0.81), major composite adverse events (arterial/venous thromboembolism, major bleed, and death; 4.2% vs. 7.9%, respectively, P = 0.07) and major bleeds (3.3% vs. 5.5%, respectively, P = 0.25) were observed. The thromboembolic event rates were 2.4% for UFH and 0.9% for LMWH. Logistic regression analysis revealed that for postoperative heparin use a Charlson comorbidity score > 1 was an independent predictor of a major bleed and that vascular, general, and major surgery were associated with non-significant trends towards an increased risk of major bleed. Conclusions: Treatment-dose LMWH, mostly in the outpatient setting, is used substantially more often than UFH as bridging therapy in patients with predominately arterial indications for OAC. Overall adverse events, including thromboembolism and bleeding, are similar for patients treated with LMWH or UFH. Postoperative heparin bridging should be used with caution in patients with multiple comorbidities and those undergoing vascular, general, and major surgery. These findings need to be confirmed using large randomized trials for specific patient groups undergoing specific procedures. Introduction Long-term oral anticoagulant (OAC) therapy is frequently and increasingly prescribed for patients at risk of arterial or venous thromboembolism, such as patients with atrial fibrillation, mechanical heart valves, or prior thromboembolic events [1]. Despite the large number of patients requiring temporary interruption of OAC for an elective surgical or invasive procedure, the management of such patients has been poorly investigated [2]. Physicians must balance the risk of thromboembolic events if OAC is discontinued (both from warfarin withdrawal and the procedure involved), and the risk of bleeding from the procedure if warfarin is continued [3]. To minimize bleeding risks, surgery can be performed once the International Normalized Ratio (INR) is below 1.5 [4]. It usually takes approximately 5 days after the last dose of warfarin to reach this level [4] and 3 days or more to reach a therapeutic INR on recommencement of warfarin [5]. Patients undergoing surgery are therefore exposed to a window of increased risk of thromboembolic events while OAC is subtherapeutic. During perioperative withdrawal of OAC, one strategy to maintain a degree of functional anticoagulation is to administer short-acting parenteral anticoagulants such as heparin while OAC therapy is subtherapeutic, a strategy called bridging therapy. This was usually accomplished by admitting patients to the hospital prior to surgery and administering intravenous unfractionated heparin (UFH) [6]. Although the safety and efficacy of low-molecular-weight heparin (LMWH) for the prophylaxis and treatment of deep vein thrombosis (DVT) and treatment of acute coronary syndromes have been well described in clinical trial settings [7-10], previous studies of LMWH bridging therapy consisted of small case series [11-14]. Only recently have observational data from large-scale, single-arm studies reported the efficacy and safety of LMWH as perioperative bridging therapy in patients that require temporary interruption of OAC [15-18]. There is currently a lack of consistent recommendations regarding heparin bridging strategies during temporary discontinuation of OAC, and a recent review of perioperative management strategies including bridging therapy highlighted the overall poor quality of the reports identified [2]. Furthermore, physician surveys reflect the variability in physicians’ practices in the administration of UFH and LMWH as bridging therapy [19, 20]. The most recent American College of Chest Physicians consensus guidelines suggest the use of preoperative and postoperative prophylactic (or higher) doses of UFH or LMWH as perioperative bridging therapy in patients considered at intermediate risk of thromboembolic events, and preoperative and postoperative full-dose UFH or LMWH in patients considered at high risk of thromboembolic events. However, these recommendations are based upon Grade 2C evidence, the weakest evidence available to the consensus panel [21]. There have been no large-scale, prospective, multicenter studies comparing practice patterns as well as the safety, efficacy, and resource utilization of LMWH and UFH as perioperative bridging therapy. REGIMEN represents the first and largest study to date conducted in North America comparing patient/procedure characteristics and clinical outcomes of patients on long-term OAC requiring temporary discontinuation and bridging therapy with UFH or LMWH for an elective procedure or surgery. Methods Study design REGIMEN was a multicenter, prospective, observational registry study involving 14 centers in the USA and Canada. The study was coordinated by the Lovelace Respiratory Research Institute in Albuquerque, NM, and was approved by the Institutional Review Boards of each participating site. All patients gave written informed consent for their data to be used in the study. Patients and treatments Consecutive patients were enrolled in the study from 1 July 2002 to 31 December 2003. The patient inclusion criteria were: age at least 18 years; undergoing a major or minor surgical procedure which necessitated temporary discontinuation of OAC; duration of OAC before the procedure of at least 3 months; and heparin (in any form) used as a bridge to OAC for at least 2 days during either or both the preoperative or postoperative period. Patients were excluded if they were enrolled in another bridging study or trial within 30 days. The study was observational: the type of bridging therapy was at the discretion of the treating doctors and was not determined by a prespecified study protocol. Data were collected on: patient demographics; indication for and length of time on OAC therapy; mean INR in the 3 months preceding the procedure; comorbidities at the time of procedure (as defined by the Charlson score – a score of 0, 1, 2, or 3 or more based upon The International Classification of Disease that is predictive of postoperative complications and hospital resource use) [22]; concomitant medication use, use of intraprocedural thrombolytics or anticoagulants, classification of risk of thromboembolism (venous or arterial), and bleeding (as defined by Beyth bleeding index [23]); the type and description of the procedure/surgery requiring cessation of OAC therapy; type of anesthesia; number of days off OAC therapy and INR values preprocedure; the type and dosage of bridging method used (UFH or LMWH in prophylactic or treatment doses); and resource utilization. Adverse events included death, arterial or venous thromboembolism, major or minor bleeds, and thrombocytopenia – for definitions see Table 1. A composite outcome was defined as any arterial or venous thromboembolism, major bleed, or death. Adverse events were adjudicated by an independent committee blinded to the specific heparin bridging strategy patients received. Table 1. Definitions of surgical procedures, arterial and venous thromboembolism, bleeding events, and thrombocytopenia Major surgery Duration > 45 min and orthopedic, cardiothoracic, vascular, or general surgical procedures Minor surgery All other surgical procedures Arterial thromboembolism Documented cardiac valvular or mural thrombus confirmed by echocardiogram Documented stroke confirmed by CT or MRI of the head or documented transient ischemic attack Documented peripheral arterial thromboembolic event confirmed by arteriography, MR angiography, spiral CT or Doppler studies Venous thromboembolism Documented DVT or PE confirmed by Doppler studies, venography or pulmonary arteriography, high probability ventilation/perfusion scan, spiral CT, or MRI Major bleed Bleeding from a critical site (retroperitoneal, intracranial, intraocular, or intraspinal bleed) ≥ 3 g dL−1 decrease in hemoglobin postprocedure Clinically apparent bleed requiring transfusion of ≥ 2 U of packed red blood cells Any bleeding resulting in new or prolonged hospitalization or surgical intervention Minor bleeding All other bleeding events Thrombocytopenia Drop in platelet count < 100 000 or > 50% of baseline with or without heparin-associated antiplatelet antibodies CT, computed tomography; DVT, deep vein thrombosis; PE, pulmonary embolism; MRI, magnetic resonance imaging. Data were collected prospectively by trained study staff at the time of enrollment, during the time of procedure or surgery, and approximately 30 days after the patient's procedure or surgery by review of the patient's medical records, telephone call to the patient, or telephone call to the treating physician. Patients were divided into two groups: patients receiving UFH alone and patients receiving LMWH alone. Two other groups also emerged: patients receiving both UFH and LMWH and patients who underwent multiple procedures. Statistical methods The final analysis compared the UFH-alone group with the LMWH-alone group. Univariate differences between the treatment groups were compared using the chi-squared test or Fisher's exact test for categorical variables using SAS software (SAS Research Institute Inc., Cary, NC, USA). A Student's t-test was used for demographic data to compare mean values of the groups. In addition, a logistic regression model was used to predict the risk of minor bleed events, major bleed events, or a composite outcome of any arterial or venous thromboembolism, major bleed, or death. Univariate predictor variables that were significant at a level of P < 0.1 were included in a multivariate model, together with the treatment group. Results Bridging treatment In total, 1077 patients were enrolled in the study. Of these, 176 patients were excluded: 123 patients did not receive the same type of heparin before and after the procedure, 41 patients underwent more than one procedure, and a further 12 received mixed types of heparin and underwent multiple procedures. Of the 53 patients undergoing multiple procedures, the majority (45) had treatment dose UFH intravenously as bridge therapy and had an interventional radiology procedure associated with cardiothoracic surgery (n = 22), a gastrointestinal procedure (n = 9), a repeat interventional radiology procedure (n = 6), or general surgery (n = 3). The analysis population therefore consisted of 901 patients: 180 receiving UFH alone and 721 receiving LMWH alone. Heparin was given at a treatment dose to 129 (72%) of the UFH group intravenously and 550 (76%) of the LMWH group subcutaneously. For the LMWH group, enoxaparin was used in 83% of patients, dalteparin in 14%, and tinzaparin in 3%. More than 95% of the LMWH treatment group received a twice-daily-dosing scheme. Patient and procedure characteristics Baseline demographics are shown in Table 2. Overall, the mean age was 65.7 years and the majority of patients (n = 671, 74.5%) had arterial indications for OAC, including a total of 246 patients (27.3%) with mechanical heart valves. Patients had multiple comorbidities as shown by a mean Charlson score of approximately 2.0 and 99 patients (11.0%) were at high bleeding risk as defined by the outpatient bleeding index. The mean INR prior to withholding warfarin was 2.5 for both groups. Patients on LMWH were significantly younger, had a lower mean Charlson score, and were less likely to be receiving OAC for arterial indications than patients on UFH (71.2% vs. 87.8%, P < 0.001), including mechanical heart valve indications (24.1% vs. 40.0%, P < 0.001). Table 2. Demographics and characteristics of bridged patients Demographic/characteristic UFH (n = 180) LMWH (n = 721) P-value Mean age (SE) 68.2 (0.91) 65.1 (0.49) <0.01 Male, n (%) 102 (56.7) 429 (59.5) 0.50 Indications for oral anticoagulation therapy: Arterial, n (%) 158 (87.8) 513 (71.2) <0.001 With mechanical valve 72 (40.0) 174 (24.1) <0.001 Without mechanical valve/with atrial fibrillation 60 (33.3) 289 (40.1) 0.10 Venous, n (%) 22 (12.2) 208 (28.8) <0.001 Mean INR (SE)* 2.5 (0.04) 2.5 (0.02) 0.74 Mean Charlson score (SD) 2.0 (0.22) 1.7 (0.10) 0.02 High bleeding risk†, n (%) 23 (12.8) 76 (10.5) 0.39 INR, international normalized ratio; LMWH, low-molecular-weight heparin; UFH, unfractionated heparin. *Information missing for some patients. †Based on outpatient bleeding index [23]. The characteristics of the procedures are shown in Table 3. Overall, 43.7% of patients (n = 394) were bridged for major surgery, and 33.3% of patients (n = 300) underwent general anesthesia. Patients on LMWH were less likely to undergo major or cardiothoracic surgery, receive intraprocedural anticoagulants or thrombolytics, or receive general anesthesia than patients receiving UFH (all P < 0.05). A significantly higher proportion of LMWH-bridged patients than UFH-bridged patients were started on OACs within 24 h of surgery (62.7% vs. 38.3%, P < 0.001). The mean number of days when OAC was not received was similar in both groups (UFH 6.0 days vs. LMWH 5.8 days, P = 0.53), while the LMWH-bridged group received heparin therapy for significantly more days than the UFH-bridged group (8.6 vs. 6.8 days, P < 0.001). Table 3. Characteristics of bridging procedures Characteristic UFH (n = 180) LMWH (n = 721) P-value Medications of interest, n (%) Antiplatelet 37 (20.6) 160 (22.2) 0.63 Other medications of interest 31 (17.2) 133 (18.4) 0.70 Anesthesia used, n (%) General 77 (42.8) 223 (30.9) <0.01 Regional 10 (5.6) 32 (4.4) – Intravenous conscious sedation 69 (38.3) 261 (36.2) 0.60 Other 23 (12.8) 198 (27.5) <0.001 Oral anticoagulants started <24 h postoperatively, n (%) 69 (38.3) 452 (62.7) <0.001 Types of surgical procedures, n (%) Orthopedic 9 (5.0) 109 (15.1) <0.001 Cardiothoracic 50 (27.8) 60 (8.3) <0.001 Interventional radiology 40 (22.2) 110 (15.3) 0.03 Urological 13 (7.2) 52 (7.2) 1.00 Gastrointestinal 15 (8.3) 148 (20.5) <0.01 Dental 2 (1.1) 49 (6.8) <0.001 General surgery 20 (11.1) 52 (7.2) 0.09 Ophthalmological 1 (0.6) 20 (2.8) 0.10 Gynecological 4 (2.2) 18 (2.5) 1.00 Dermatological 3 (1.7) 33 (4.6) 0.09 Vascular surgery 4 (2.2) 11 (1.5) 0.52 ENT (ear, nose, throat) 5 (2.8) 15 (2.1) 0.57 Other 14 (7.8) 44 (6.1) 0.40 Intraprocedural anticoagulants or thrombolytics used 33 (18.3) 44 (6.1) 45 min* 116 (69.0) 278 (41.6) <0.001 Length of hospital stay, n (%) Outpatient or <24 h 11 (6.1) 458 (63.5) <0.001 Hospitalized, mean length of stay (SE) 10.3 (0.53) 4.6 (0.35) <0.001 Mean days off oral anticoagulant therapy (SE) 6.0 (0.30) 5.8 (0.11) 0.53 Mean days on heparin therapy (SE) 6.8 (0.30) 8.6 (0.19) <0.001 Patients receiving a postop heparin dose, n (%) 164 (91.1) 668 (92.6) 0.49 LMWH, low-molecular-weight heparin; UFH, unfractionated heparin. *Data missing for some patients. Overall, the majority of patients in the LMWH-bridged group had the procedure performed completely on an outpatient basis or with hospitalization of < 24 h compared with the UFH-bridged group (63.6% vs. 6.1%, P < 0.001) (Table 3). Of patients who were hospitalized, the LMWH-bridged group had a significantly decreased length of hospital stay compared with the UFH-bridged group (4.6 vs. 10.3 days, P < 0.001). Outcomes Of those 901 patients who received bridging therapy, 832 received a postoperative heparin dose: 164 (91.1%) in the UFH group and 668 (92.6%) in the LMWH group (P = 0.49) (Table 3). There were no significant differences in the number of adverse events experienced by UFH- and LMWH-bridged patients (17.1% vs. 16.2%, respectively, P = 0.81), with the majority of adverse events being minor bleeds (Table 4). The overall composite outcome consisting of thromboembolic complications, major bleed, or death was 7.9% in the UFH group and 4.2% in the LMWH group (P = 0.07). Table 4. Adverse events in bridged patients Adverse event UFH (n = 164) LMWH (n = 668) P-value Any adverse event, n (%) 28 (17.1%) 108 (16.2%) 0.81 Arterial/venous thromboembolism, major bleed, or death 13 (7.9%) 28 (4.2%) 0.07 Adverse events, n (%) Arterial thromboembolism 4* (2.4) 4† (0.6) – Venous thromboembolism 0 (0) 2‡ (0.3) – Major bleed 9 (5.5) 22 (3.3) 0.25 Minor bleed 15 (9.1) 80 (12.0) 0.34 Thrombocytopenia 2 (1.2) 3 (0.4) – Death 2 (1.2) 4 (0.6) – LMWH, low-molecular-weight heparin; UFH, unfractionated heparin. *One cardiac valvular or mural thrombosis, one intracranial event, one transient ischemic attack, one peripheral arterial event. †Two intracranial events and two transient ischemic attacks. ‡Two deep vein thromboses. Thromboembolic event rates in the UFH group were as follows: 2.4% (4/164) of patients experienced an arterial thromboembolic event (one cardiac valvular or mural thrombosis, one intracranial event, one transient ischemic attack, and one peripheral arterial event), and no patients had venous thromboembolism. In the LMWH group, 0.6% (4/668) of patients experienced an arterial thromboembolic event (two intracranial events and two transient ischemic attacks), and 0.3% (2/668) of patients had a DVT. The arterial thromboembolic event rates did not change appreciably in either group if expressed in patients solely with arterial indications for OAC. Major bleed rates were similar in both the UFH and LMWH groups (5.5% vs. 3.3%, respectively, P = 0.25). There were two (1.2%) deaths in the UFH group (cardiac arrest, adult respiratory distress syndrome) and four (0.6%) deaths in the LMWH group (septic shock, intracerebral hemorrhage possibly due to an aneurysm, myocardial infarction, and cardiopulmonary collapse due to congestive heart failure in a patient with an implantable defibrillator). Major bleeds occurred in 4.9% of patients undergoing major surgery, and 3.1% in patients undergoing minor surgery, a difference that was not statistically significant (P = 0.19). In 732 patients who received postoperative heparin (either UFH or LMWH) and had timing information, similar rates of major bleeds were observed between those who received a first heparin dose within 12 h postoperatively and those who did not (4.4% vs. 3.8%, P = 0.68). There were no differences in the major bleed rates between treatment and prophylactic doses of heparin given postoperatively (data not shown). The timing of postoperative major bleeds, expressed as the mean number of days that bleeding occurred after surgery, did not vary significantly between the UFH and LMWH groups (7.3 vs. 8.2 days, P = 0.77). No bleed events occurred preoperatively. Among the 31 patients who received a postoperative dose of heparin and experienced a major bleed event, 90.3% had an INR measurement. The mean INR value was 1.8 (range 1.0–3.3). Both the UFH and LMWH groups also experienced low rates of thrombocytopenia (1.2% vs. 0.4%, respectively). Univariate logistic regression analysis showed that a Charlson score >1, general surgery, and general anesthesia were all significant univariate predictors of a major bleed event, while vascular surgery and major surgery were associated with trends approaching significance. Multivariate analyses A Charlson score of >1 was a significant independent predictor of a major bleed [odds ratio (OR) 2.24, 95% confidence interval (CI) 1.02–4.93; Table 5]. General, vascular, and any major surgery were all associated with trends as predictors of a major bleed (Table 5). A trend toward a reduction of a major bleed event was observed in patients receiving LMWH compared with those receiving UFH (Table 5). Table 5. Logistic regression analysis for major bleed (n = 774) Independent variable Reference group Odds ratio 95% confidence interval Vascular surgery Not vascular surgery 4.72 0.92–24.09 General surgery Not general surgery 2.25 0.78–6.53 Charlson score > 1 Charlson score≤ 1 2.24 1.02–4.93 Procedure duration > 45 min Procedure duration ≤ 45 min 1.37 0.59–3.18 LMWH postoperatively UFH postoperatively 0.76 0.32–1.81 LMWH, low-molecular-weight heparin; UFH, unfractionated heparin. Numbers in bold are significant at the level of P ≤ 0.05. Discussion This is the first large, prospective, multicenter registry completed in North America assessing patient/procedure characteristics and clinical outcomes with either UFH or LMWH as heparin bridging therapy for patients on chronic OAC requiring temporary interruption for an elective procedure or surgery. The study included patients with both arterial (atrial fibrillation and mechanical heart valve) and venous indications for OAC, patients receiving either prophylactic or treatment doses of heparin (either UFH or LMWH), and patients undergoing both major and minor procedures. The study represents a wider scope of heparin bridging strategies, patient populations, and procedures than previous clinical studies or registries of heparin bridging therapy [6, 11-18]. The present registry describes several key characteristics of patients who received bridging therapy and the procedures for which bridging is considered. This population represented an elderly (mean age > 65 years) and comorbid group, with the majority having arterial indications for OAC. Furthermore, approximately 11% of patients were at high risk of bleeding and about 33% and 45% of the population underwent general anesthesia and major surgery, respectively. This study found that heparin is used preferentially in treatment doses over prophylactic doses as bridging therapy, and that LMWH, usually in a twice-daily-dosing scheme, is more widely used than UFH as bridging therapy. These data correspond well with physician surveys conducted in 1998 and 2003 that reveal changes in practice patterns with bridging therapy, with greater use of therapeutic-dose LMWH over UFH, and increased use of bridging therapy in patients considered at intermediate or high risk of thromboembolism, such as those with atrial fibrillation [20, 24]. The LMWH group was younger and had a lower comorbidity index, underwent less cardiothoracic or major surgery, was less likely to receive general anesthesia, and was less likely to receive intraprocedural anticoagulants or thrombolytics than the UFH group. This may indicate that cardiothoracic, vascular, and general surgeons, as well as cardiologists, still prefer a very short-acting periprocedural anticoagulant, especially with procedures that incur a high bleeding risk. Nearly one-quarter of patients bridged with LMWH had mechanical heart valves. This is also the first large, multicenter, bridging study to directly compare resource utilization between LMWH- and UFH-bridged groups. The study found that nearly 65% of patients bridged with LMWH underwent their procedure completely in an outpatient setting or with < 24 h of hospitalization, compared with approximately 6% of patients bridged with UFH. In addition, of patients requiring hospitalization, the LMWH-bridged group had a 55% reduction in length of hospital stay compared with the UFH-bridged group. This study adds to the emerging body of pharmacoeconomic data showing that significant cost savings can be achieved using LMWH as bridging therapy, mostly in an outpatient setting, compared with intravenous UFH used in-hospital [14, 25, 26]. In terms of outcomes, the observed rates of thromboembolism (2.4% for UFH, 0.9% for LMWH) and major bleed (5.5% for UFH, 3.3% for LMWH) correspond well with events reported in recent, large, prospective studies of bridging therapy, with thromboembolic event rates of 0.6–3.6% and major bleed rates of 0.9–6.7% [15-18]. If restricted to suspected cardioembolism, the event rate with the LMWH group was 0.6%, which is consistent with previous studies of bridging therapy with LMWH in patients at increased risk of arterial thromboembolism (range 0–0.9%) [13, 15-18]. Of note, the overall major bleed rate of 3.3% for LMWH is more in line with previous studies using a twice-daily treatment dose of LMWH bridging therapy [13, 15, 18] as recent clinical studies in bridging therapy suggest an increase in postoperative major bleeds if a once-daily treatment dose of LMWH is used, especially for major procedures [16, 17]. A post hoc analysis revealed a trend toward an increase in major bleed rates in patients receiving a postoperative dose of heparin and undergoing a major vs. a minor procedure (4.9% vs. 3.1%). We tested for, but did not find differences in major bleed rates when comp