Article Clinical Impact of Bleeding in Patients Taking Oral Anticoagulant Therapy for Venous Thromboembolism A Meta-Analysis Lori-Ann Linkins, MD, FRCP(C); Peter T. Choi, MD, MSc, FRCP(C); and James D. Douketis, MD, FRCP(C)

Background: Clinicians should consider the clinical impact of anticoagulant-related bleeding when deciding on the duration of anticoagulant therapy in patients with venous thromboembolism.

Data Synthesis:

Controlled Trial Registry, thromboembolism experts, and reference lists; English-language literature only.

The authors analyzed 33 studies involving 4374 patient-years of oral anticoagulant therapy. For all patients, the case-fatality rate of major bleeding was 13.4% (95% CI, 9.4% to 17.4%) and the rate of intracranial bleeding was 1.15 per 100 patient-years (CI, 1.14 to 1.16 per 100 patient-years). For patients who received anticoagulant therapy for more than 3 months, the case-fatality rate of major bleeding was 9.1% (CI, 2.5% to 21.7%), and the rate of intracranial bleeding was 0.65 per 100 patient-years (CI, 0.63 to 0.68 per 100 patient-years) after the initial 3 months of anticoagulation.

Study Selection: Randomized, controlled trials and prospective cohort studies that investigated patients with venous thromboembolism who received oral anticoagulant therapy (target international normalized ratio, 2.0 to 3.0) for at least 3 months and that reported major bleeding and death as primary study outcomes.

Conclusion: The clinical impact of anticoagulant-related major bleeding in patients with venous thromboembolism is considerable, and clinicians should take this into account when deciding whether to continue long-term oral anticoagulant therapy in an individual patient.

Purpose:

To provide reliable estimates of the clinical impact of anticoagulant-related bleeding, defined as the case-fatality rate of major bleeding and the risk for intracranial bleeding.

Data Sources: MEDLINE (January 1989 to May 2003), Cochrane

Data Extraction:

Two reviewers independently extracted data on the number of anticoagulant-related major and intracranial bleeding episodes and on whether these events were fatal or nonfatal.

I

n patients with venous thromboembolism, which includes deep venous thrombosis and pulmonary embolism, the duration of oral anticoagulant therapy is driven mainly by the need to prevent recurrent disease, whereas the bleeding risks of anticoagulation are considered less important. This approach may be due to the perception that the clinical impact of recurrent venous thromboembolism, which manifests as fatal pulmonary embolism in its most severe form, is greater than the clinical impact of anticoagulant-related bleeding, which is generally considered a treatable and self-limiting complication (1, 2). Furthermore, although the risk and clinical impact of recurrent venous thromboembolism have been widely investigated (3–5), reliable estimates of the risk and clinical impact of anticoagulant-related bleeding are lacking. Determining the consequences of anticoagulant-related bleeding is important for clinicians weighing the risks and benefits of extended-duration anticoagulation, particularly in the subgroup of patients with venous thromboembolism in whom the optimal duration of anticoagulant therapy is unclear. These patients include those with unprovoked (or idiopathic) venous thromboembolism (6 –9) and heterozygous carriers of the factor V Leiden and prothrombin G20210A mutations (10). Patients with deficiencies of protein S or protein C are possibly included in this group since their need for life-long anticoagulation has recently been questioned (11). Collectively, such patients make up about half (6, 12) of an estimated 450 000 pa-

Ann Intern Med. 2003;139:893-900. For author affiliations, see end of text.

www.annals.org

tients in whom symptomatic venous thromboembolism is diagnosed in the United States per year (13–15). It is not feasible to determine the clinical impact of anticoagulant-related bleeding from single clinical trials involving patients with venous thromboembolism because major bleeding events are infrequent, occurring in about 2% to 3% of patients per year (6 –9, 16, 17). Previous studies investigating the consequences of anticoagulantrelated bleeding in patients who were receiving oral anticoagulant therapy are limited because these studies did not focus on patients with venous thromboembolism (18 –22), investigated only the clinical impact of intracranial bleeding (23, 24), or did not investigate the case-fatality rate of major bleeding (18 –23, 25–27). We therefore performed a meta-analysis of studies involving patients with venous thromboembolism who received at least 3 months of oral anticoagulant therapy with a target international normalized ratio (INR) of 2.0 to 3.0. Our objective was to obtain reliable estimates of the clinical impact of anticoagulant-related bleeding, defined as the case-fatality rate of major bleeding and the risk for intracranial bleeding.

METHODS Study Identification

We searched the MEDLINE and Cochrane Controlled Trials Registry electronic databases (English-language articles © 2003 American College of Physicians 893

Article

Clinical Impact of Bleeding during Oral Anticoagulant Therapy

Context Clinical impact of bleeding from long-term oral anticoagulation for venous thromboembolism is not well quantified.

Contribution This summary of prospective studies involving patients with venous thromboembolism showed that in studies with more than 6 months of follow-up, the rates of major and intracranial bleeding were 2.06% and 1.48% during the initial 3 months of anticoagulation. After 3 months, the rates of major and intracranial bleeding per 100 patient-years were 2.74 and 0.65, respectively. The casefatality rate of major bleeding before and after 3 months of anticoagulation was 9.3% (95% CI, 3.1% to 20.3%) and 9.1% (CI, 2.5% to 21.7%), respectively.

Implications Extended-duration oral anticoagulation for venous thromboembolism is associated with significant risk for bleeding. –The Editors

only) to identify prospective studies published from January 1989 through May 2003 that involved patients with venous thromboembolism who were receiving oral anticoagulant therapy. We excluded articles published before 1989 because before that year, most clinical centers were still using prothrombin time instead of INR to monitor the intensity of oral anticoagulant therapy (28). (The introduction of the INR standardized the reporting of prothrombin time by taking into account the wide range of responsiveness of thromboplastin reagents to decreases in vitamin K– dependent coagulation factors.) The key words used for this search strategy were deep vein thrombosis, pulmonary embolism, warfarin, coumarin, and treatment. We also manually reviewed relevant bibliographies and conferred with thrombosis experts to identify articles that may have been overlooked by the database search (29). Study Selection

Studies were included if they satisfied the following 4 criteria: 1) The study was a randomized, controlled trial or prospective cohort study involving consecutive patients with acute venous thromboembolism; 2) the study confirmed venous thromboembolism by compression ultrasonography, venography, ventilation–perfusion lung scanning, or pulmonary angiography; 3) the patients received a coumarin derivative for at least 3 months, administered to achieve a target INR of 2.0 to 3.0; and 4) bleeding events were classified as major, minor, or fatal. Studies were excluded if patients received an antiplatelet agent in addition to a coumarin derivative. Studies that enrolled fewer than 100 patients were also excluded since skewed results that misrepresent a treatment-related risk or benefit are more likely in small studies (30). Two authors independently reviewed eligible studies 894 2 December 2003 Annals of Internal Medicine Volume 139 • Number 11

to assess suitability for inclusion and accuracy of data extraction. Agreement was assessed by the weighted ␬ statistic (31), and disagreements were resolved by consensus. When relevant data from a study were missing or unclear, we contacted the study investigators to provide or clarify the data. Data Extraction

For each study, we extracted data on the number of major and intracranial bleeding episodes that occurred during anticoagulant therapy and on whether such events were nonfatal or fatal. We defined major bleeding episodes as those that were clinically overt and were associated with at least 1 of the following 4 characteristics (32): 1) required hospitalization, 2) required transfusion of at least 2 units of packed red blood cells, 3) were intracranial or retroperitoneal or involved a body cavity, or 4) were fatal. Deaths that were reported as “associated with bleeding” were counted as fatal bleeding episodes. For studies with more than 3 months of on-treatment follow-up, data were extracted on whether bleeding episodes occurred during the initial 3 months of anticoagulation or after this period. Statistical Analysis

We pooled the results of individual studies to determine weighted point estimates and associated 95% CIs for the rate of major, fatal, and intracranial bleeding using Excel software, version 2000 (Microsoft Corp., Redmond, Washington), and the single-proportion random-effects model (33). The weighting of outcomes adjusts for differences in sample size and anticoagulant duration in individual studies and is based on the reciprocal of the variance of bleeding rates and the between-study variance across studies. The case-fatality rate of major bleeding was expressed as a percentage (the number of fatal bleeding episodes divided by the number of fatal plus nonfatal bleeding episodes). Bleeding outcomes were allocated according to the intention-to-treat principle. For analyses involving all studies, findings were expressed as bleeding rates per 100 patient-years of anticoagulant therapy because we combined studies that had variable durations of treatment. For analyses limited to randomized, controlled trials in which patients received more than 3 months of anticoagulation, outcomes are reported separately for 1) the initial 3 months of anticoagulation and 2) after the initial 3 months of anticoagulation. For pooled analyses of outcomes during the initial 3 months of anticoagulation, findings are expressed as 3-month bleeding rates (percentages). For pooled analyses of outcomes after the initial 3 months of anticoagulation, outcomes are expressed as bleeding rates per 100 patientyears of treatment because we pooled variable follow-up durations. Role of the Funding Sources

The funding sources had no role in the collection, analysis, or interpretation of the data or in the decision to submit the manuscript for publication. www.annals.org

Clinical Impact of Bleeding during Oral Anticoagulant Therapy Figure. Literature search and selection.

DATA SYNTHESIS Studies

Of 124 potentially eligible studies, 33 studies—29 randomized, controlled trials (6 –9, 34 –58) and 4 prospective cohort studies (59 – 62)—were included in the analysis (Figure). The interrater agreement for study eligibility and data extraction was good (weighted ␬ ⫽ 0.79 and 0.71, respectively [31]). For 3 studies (34, 51, 55), data on the number of intracranial bleeding episodes were not available despite attempts to contact the study investigators. Consequently, relevant data from these studies were excluded from the analysis of intracranial bleeding risk. Although we did not formally assess the quality of the included studies, our selection criteria were designed to limit our analysis to studies with a clearly defined patient group (those with objectively confirmed venous thromboembolism) who underwent a standardized intervention (oral anticoagulant therapy with a target INR of 2.0 to 3.0 for ⱖ3 months) and follow-up to detect the same outcome (fatal and nonfatal major bleeding episodes). Furthermore, the completeness of patient follow-up was acceptable—26 studies had no loss to follow-up and 7 had 5% or less loss to follow-up (6, 38 – 41, 46, 51) (Table 1). Differences in study design and patient characteristics across studies might have influenced outcome rates. In terms of patients’ baseline risk for bleeding across studies, all studies excluded patients with active bleeding, severe hepatic failure, and severe renal failure (the definition of renal failure differed slightly among studies). Seventeen of the studies made general statements that they excluded patients with “disorders that contraindicated anticoagulant therapy” or patients who had “a contraindication to anticoagulant treatment” (34 –36, 38 – 40, 49, 50, 53, 55, 60), whereas other studies outlined these disorders more specifwww.annals.org

Article

ically (for example, uncontrolled hypertension, history of gastrointestinal ulcer or stroke, recent surgery, and thrombocytopenia). Three studies excluded patients with a malignant condition (8, 9, 52), and 4 studies excluded patients with a life expectancy of less than 3 months (40, 48, 54, 58). Eleven studies prohibited or strongly discouraged use of platelet-inhibiting or nonsteroidal anti-inflammatory drugs (6, 7, 43, 44, 47, 50, 54 –58). In terms of study design, there were potential discrepancies in the initiation of anticoagulant therapy across studies: The initial dose of warfarin (the oral anticoagulant used by most studies) was 10 mg in 7 studies (39, 43– 45, 57–59) and 5 mg in 2 studies (41, 53) and was not reported in 20 studies (6 –9, 34, 36, 38, 40, 42, 46, 48, 49, 51, 52, 54 –56, 60 – 62). However, randomized, controlled trials that have compared initiation of oral anticoagulant therapy with 10 mg or 5 mg have found no differences in bleeding rates with either dose (63, 64). The definition of major bleeding differed across studies; this variation is expected since there is no reference standard definition for major bleeding (65, 66). Twenty-one studies used our prespecified definition for major bleeding (8, 9, 34, 35, 37, 38, 41, 42, 45– 48, 50, 51, 53–56, 60 – 62). In 7 studies (39, 40, 43, 44, 52, 57, 58), the definition of major bleeding slightly differed—intra-articular bleeding was considered major bleeding or retroperitoneal hemorrhage was not explicitly specified as a major bleeding episode. In 5 studies (6, 7, 36, 49, 59), major bleeding required a different decrease in hemoglobin level or number of units of packed red blood cells than stated in our definition. Overall, when studies that used our definition of major bleeding were pooled and compared with studies that had a slightly different definition of major bleeding, the bleeding rates did not significantly differ; this finding supports our pooling of major bleeding episodes across studies. Clinical Impact of Bleeding Entire Period of Anticoagulation

Our analysis included 10 757 patients who received 4374 patient-years of anticoagulant therapy. Table 1 shows the bleeding rates for individual studies. Major bleeding occurred at a rate of 7.22 per 100 patient-years (95% CI, 7.19 to 7.24), and fatal bleeding occurred at a rate of 1.31 per 100 patient-years (CI, 1.30 to 1.32). Intracranial bleeding occurred at a rate of 1.15 per 100 patient-years (CI, 1.14 to 1.16) and accounted for 8.7% of all major bleeding episodes. Eleven of 24 intracranial bleeding episodes were fatal. A total of 276 major bleeding episodes occurred in patients receiving anticoagulant therapy, of which 37 were fatal; the resulting case-fatality rate for major bleeding was 13.4% (CI, 9.4% to 17.4%). During and after the Initial 3 Months of Anticoagulation

Nine studies involved 2422 patients who received anticoagulant therapy for more than 3 months (6 –9, 35, 37, 40, 42, 52). Table 2 shows bleeding outcomes for these 2 December 2003 Annals of Internal Medicine Volume 139 • Number 11 895

Article

Clinical Impact of Bleeding during Oral Anticoagulant Therapy

Table 1. Major Bleeding Events in Patients Receiving Oral Anticoagulant Therapy* Study, Year (Reference)

Study Type

Hull et al., 1990 (44) Prandoni et al., 1992 (53) Lopaciuk et al., 1992 (49) Brandjes et al., 1992 (37) Hull et al., 1992 (57) Pini et al., 1994 (58) Levine et al., 1995 (48) Schulman et al., 1995 (6) Lindmarker et al., 1996 (59) Koopman et al., 1996 (46) Das, 1996 (39) Bounameaux et al., 1997 (36) Columbus Investigators, 1997 (34) Schulman et al., 1997 (7) Simonneau et al., 1997 (54) Monreal et al., 1998 (60) Wells et al., 1998 (61) Charbonnier et al., 1998 (38) Decousus et al., 1998 (40) Kovacs et al., 1998 (45) Leroyer et al., 1998 (47) Lopaciuk et al., 1999 (50) Gonzalez-Fajardo et al., 1999 (41) Kearon et al., 1999 (8) Harenberg et al., 2000 (42) Boccalon et al., 2000 (35) Hull et al., 2000 (43) Rembrandt Investigators, 2000 (55) Kovacs et al., 2000 (62) Agnelli et al., 2001 (9) Merli et al., 2001 (51) Pinede et al., 2001 (52) Breddin et al., 2001 (56) Total

RCT RCT RCT RCT RCT RCT RCT RCT Cohort RCT RCT RCT RCT RCT RCT Cohort Cohort RCT RCT RCT RCT RCT RCT RCT RCT RCT RCT RCT Cohort RCT RCT RCT RCT

Patients Receiving Oral Anticoagulants

Duration of Follow-up

Major Bleeding Episodes

CaseFatality Rate

Intracranial Bleeding Episodes

Loss to Follow-up

n

mo

n (%)

%

n

n (%)

199 170 146 120 432 94 301 454 434 400 55 47 1021 227 612 244 194 651 400 111 223 95 80 79 538 201 200 119 108 134 900 631 1137 10 757

3 3 3 6 3 3 3 6 3 3 3 3 3 6 or 48 3 3 or 6 3 3 24 3 3 3 3 24 6 6 3 3 3 12 3 3 or 6 3

16 (8) 8 (5) 0 5 (4) 17 (4) 3 (3) 1 (0.3) 5 (1) 4 (0.9) 5 (1) 0 0 28 (3) 11 (5) 18 (3) 2 (0.8) 4 (2) 17 (3) 39 (10) 3 (3) 7 (3) 1 (1) 2 (3) 3 (4) 20 (4) 4 (2) 6 (3) 3 (3) 3 (2) 4 (3) 15 (2) 15 (2) 7 (0.6) 276

6 13 0 0 18 0 100 0 25 40 0 0 7 18 17 0 0 29 8 0 29 0 0 0 15 0 0 0 0 0 47 7 0

0 1 0 0 2 1 1 2 2 1 0 0 NA 1 1 1 0 1 3 0 2 0 0 0 3 0 1 NA 0 0 NA 1 0 24

0 0 0 0 0 0 0 21 (5) 0 4 (1) 3 (5) 0 0 0 0 0 0 1 (0.2) 4 (1) 0 0 0 3 (4) 0 0 0 0 0 0 0 2 (0.2) 0 0 38

* NA ⫽ not available; RCT ⫽ randomized, controlled trial.

studies. Anticoagulation lasted for 6 months in 5 studies (6, 35, 37, 42, 52), 1 year in 1 study (9), and 2 or more years in 3 studies (7, 8, 40). A total of 2422 patients received 606 patient-years of treatment during the initial 3 months of anticoagulant therapy and 1748 patient-years of anticoagulant therapy after the initial 3 months of treatment. Table 3 summarizes pooled findings for these studies. During the initial 3 months of anticoagulation, the rate of major bleeding was 2.06% (CI, 2.04% to 2.08%) and the rate of fatal bleeding was 0.37% (CI, 0.36% to 0.38%). After the first 3 months, the rate of major bleeding was 2.74 per 100 patient-years (CI, 2.71 to 2.77) and the rate of fatal bleeding was 0.63 per 100 patient-years (CI, 0.61 to 0.65). Four intracranial bleeding episodes occurred during the initial 3 months of anticoagulant therapy, of which 2 were fatal. After the first 3 months, 5 intracranial bleeding episodes occurred, of which 1 was fatal. The rate of intracranial bleeding was 1.48% (CI, 1.40% to 1.56%) and 0.65 per 100 patient-years (CI, 0.63 to 0.68) for these 2 time periods, respectively. The timing of 1 nonfatal intracranial bleeding episode was unclear and thus was not included in this analysis. 896 2 December 2003 Annals of Internal Medicine Volume 139 • Number 11

During the initial 3 months of anticoagulation, 54 major bleeding episodes occurred, of which 5 were fatal; the resulting case-fatality rate of major bleeding was 9.3% (CI, 3.1% to 20.3%). After the first 3 months of anticoagulation, 44 major bleeding episodes occurred, of which 4 were fatal (case-fatality rate of major bleeding, 9.1% [CI, 2.5% to 21.7%]).

DISCUSSION To our knowledge, this is the first study to determine the clinical impact of anticoagulant-related bleeding in patients with venous thromboembolism who are receiving oral anticoagulant therapy. Overall, the case-fatality rate of major bleeding was 13.4% and the risk for intracranial bleeding was 1.1 per 100 patient-years. Thus, the clinical impact of anticoagulant-related major bleeding is considerable—about 1 in 7 bleeding episodes are fatal or intracranial—and may be greater than previously perceived. After the initial 3 months of anticoagulant therapy, the casefatality rate of major bleeding was 9.1% and the risk for intracranial bleeding was 0.65 per 100 patient-years in pawww.annals.org

Clinical Impact of Bleeding during Oral Anticoagulant Therapy

Article

Table 2. Bleeding Events in Patients Receiving Oral Anticoagulant Therapy for More than 3 Months in Randomized, Controlled Trials* Study, Year (Reference)

Duration of Follow-up

Patients Receiving Oral Anticoagulants

Major Bleeding

3 mo

Fatal Bleeding Episode 3 mo

mo

4OOOOOOOOOOOOOOOO n OOOOOOOOOOOOOOOO3

6 6 6 or 48 24 24 6 6 12 6

120 454 227 400 79 538 201 134 269 2422

5 3 3 21 0 17 4 0 1 54

0 2 8 18 3 3 0 4 6 44

0 0 0 2 0 2 0 0 1 5

0 0 2 1 0 1 0 0 0 4

Intracranial Bleeding Episode (Fatal) 3 mo

n (n) 0 1 (0) 1 (1) 0 0 2 (1) 0 0 NA 4 (2)

0 1 (0) 0 (0) 3 (0) 0 1 (1) 0 0 NA 5 (1)

* NA ⫽ not available.

tients who received extended-duration anticoagulation. These findings are likely to be valid because the source data for our pooled analyses were derived from well-designed clinical trials in which major bleeding and death were primary study outcomes. Consequently, these outcomes were probably documented reliably. Additional findings from our study are noteworthy. First, there appears to be a front-loading of major bleeding episodes shortly after the initiation of anticoagulant therapy. As many major bleeding episodes occurred during the initial 3 months of anticoagulation (2.06 per 100 patientyears) as during the entire year after this period (2.74 per 100 patient-years). Other investigators (21, 26) have observed this apparent clustering of bleeding events at the start of anticoagulant therapy; in a 3-month study of 1021 patients with venous thromboembolism, 3 of 4 major bleeding episodes occurred during the initial 3 weeks of treatment (12). These findings are consistent with the premise that patients with an underlying predisposition to bleeding are more likely to develop this complication soon after the start of anticoagulant therapy. Another noteworthy finding is the lower proportion of fatal bleeding episodes due to intracranial bleeding compared with other studies. In our analysis, 30% (11 of 37) of fatal bleeding episodes were due to intracranial bleeding, although the details on 8 fatal bleeding episodes were unavailable and up to 51% of fatal bleeding episodes may have been intracranial. By comparison, Landefeld and Beyth (19) and Palareti and colleagues (67) reported that fatal bleeding episodes were due to intracranial bleeding in 70% (7 of 10) and 100% (5 of 5) of cases, respectively. Furthermore, in a pooled analysis of 23 studies, Levine and colleagues reported that 85% (45 of 53) of fatal bleeding episodes were intracranial (65). The most likely explanation for these discrepant findings is different patient characteristics. Previous studies included patients with chronic atrial fibrillation or other cardiac conditions who were receiving oral anticoagulant therapy for stroke prevention. www.annals.org

Such patients, who were excluded from our analysis, are more likely to develop intracranial bleeding as a manifestation of anticoagulant-related bleeding than patients with venous thromboembolism (66, 68). Finally, we found a high case-fatality rate associated with intracranial bleeding—11 of 24 intracranial bleeding episodes were fatal. This finding is consistent with the 45% (CI, 23% to 67%) case-fatality rate in a retrospective cohort study of 20 patients with oral anticoagulant–associated intracranial bleeding (24). Our study has some potential limitations. First, we acknowledge that meta-analysis is retrospective research. Thus, we could not perform a time-to-bleeding analysis, which would have provided the most accurate estimate of bleeding rates over fixed time periods. However, we could determine whether bleeding occurred during or after the initial 3 months of anticoagulant therapy, and this differentiation is relevant for guiding decisions on the safety of long-term oral anticoagulant therapy. Second, the criteria for major bleeding varied slightly across studies because there is no standardized definition for major bleeding (65, 66). However, although we could not apply our definition to each patient with a major bleeding episode, all bleeding events included in our analysis were probably clinically important since they were fatal, involved a critical organ or body cavity, or required transfusion of blood products or Table 3. Clinical Effect of Anticoagulant-Related Bleeding Time Period of Anticoagulant Therapy

Case-Fatality Rate of Major Bleeding (95% CI), %

Rate of Intracranial Bleeding (95% CI)

Entire period of anticoagulant therapy Initial 3 mo of anticoagulant therapy After initial 3 mo of anticoagulant therapy

13.4 (9.4–17.4)

1.15 (1.14–1.16) per 100 patient-years 1.48% (1.40%–1.56%)

9.3 (3.1–20.3) 9.1 (2.5–21.7)

0.65 (0.63–0.68) per 100 patient-years

2 December 2003 Annals of Internal Medicine Volume 139 • Number 11 897

Article

Clinical Impact of Bleeding during Oral Anticoagulant Therapy

hospitalization. Third, because our findings are based on data from clinical trials, our findings may not be generalizable to “real world” patients since clinical trials may exclude sicker patients. However, because our estimates of bleeding rates are consistent with those found in nonrandomized cohort studies that enrolled consecutive patients, our findings are probably consistent with the typical “real world” patient with venous thromboembolism who is considered to be at average risk for bleeding (61, 62). Our results provide clinicians with a framework in which to consider the consequences of anticoagulant-related bleeding and to weigh these risks against the benefits of long-term anticoagulation in patients with venous thromboembolism. Although every case must be considered individually, this framework is most relevant to patients with unprovoked (or idiopathic) venous thromboembolism who would be considered to be at average risk for bleeding. For example, in a typical patient with a first, unprovoked episode of deep venous thrombosis, the risk for disease recurrence after the completion of 6 months of oral anticoagulant therapy is about 7% per year (5–9, 16, 17), with a case-fatality rate of 5% (3). For the same patient, the risk for anticoagulant-related major bleeding is about 2% per year (6 –9, 17, 67), assuming an average risk for bleeding, with a case-fatality rate of about 9% as determined by our analysis. Consequently, the risk for fatal venous thromboembolism without extended anticoagulation (annual rate, 0.35%) is greater than the risk for fatal bleeding (annual rate, 0.18%) associated with anticoagulation that extends beyond 6 months. Therefore, in patients with clinical factors that confer a risk for anticoagulant-related major bleeding of 4% per year (annual rate, 0.36%) or more, the bleeding risks appear to be as great as or may outweigh the benefits of extended anticoagulant therapy. In summary, the clinical impact of anticoagulantrelated major bleeding in patients with venous thromboembolism is considerable. Clinicians should take this into account when deciding whether to maintain an individual patient on long-term oral anticoagulant therapy. From McMaster University, Hamilton, Ontario, Canada; and University of British Columbia, Vancouver, British Columbia, Canada. Acknowledgments: The authors thank Drs. Clive Kearon, Mark

Crowther, Lehana Thabane, and Jack Hirsh for their helpful reviews of the manuscript. Grant Support: Dr. Linkins is the recipient of a Fellowship Award from the Heart & Stroke Scientific Research Corporation of Canada/AstraZeneca Canada, Inc. Dr. Douketis is a recipient of a Research Scholarship from the Heart and Stroke Foundation of Canada. Potential Financial Conflicts of Interest: None disclosed. Requests for Single Reprints: James D. Douketis, MD, FRCP(C), St.

Joseph’s Healthcare, Room F-541, 50 Charlton Avenue East, Hamilton, Ontario L8N 4A6, Canada; e-mail, [email protected]. 898 2 December 2003 Annals of Internal Medicine Volume 139 • Number 11

Current author addresses and author contributions are available at www .annals.org.

References 1. Devereaux PJ, Anderson DR, Gardner MJ, Putnam W, Flowerdew GJ, Brownell BF, et al. Differences between perspectives of physicians and patients on anticoagulation in patients with atrial fibrillation: observational study. BMJ. 2001;323:1218-22. [PMID: 11719412] 2. Douketis JD, Crowther MA, Cherian SS, Kearon C. Physician preferences for perioperative anticoagulation in patients with a mechanical heart valve who are undergoing elective noncardiac surgery. Chest. 1999;116:1240-6. [PMID: 10559081] 3. Douketis JD, Kearon C, Bates S, Duku EK, Ginsberg JS. Risk of fatal pulmonary embolism in patients with treated venous thromboembolism. JAMA. 1998;279:458-62. [PMID: 9466640] 4. Anderson FA Jr, Wheeler HB, Goldberg RJ, Hosmer DW, Patwardhan NA, Jovanovic B, et al. A population-based perspective of the hospital incidence and case-fatality rates of deep vein thrombosis and pulmonary embolism. The Worcester DVT Study. Arch Intern Med. 1991;151:933-8. [PMID: 2025141] 5. Prandoni P, Lensing AW, Cogo A, Cuppini S, Villalta S, Carta M, et al. The long-term clinical course of acute deep venous thrombosis. Ann Intern Med. 1996;125:1-7. [PMID: 8644983] 6. Schulman S, Rhedin AS, Lindmarker P, Carlsson A, Larfars G, Nicol P, et al. A comparison of six weeks with six months of oral anticoagulant therapy after a first episode of venous thromboembolism. Duration of Anticoagulation Trial Study Group. N Engl J Med. 1995;332:1661-5. [PMID: 7760866] 7. Schulman S, Granqvist S, Holmstrom M, Carlsson A, Lindmarker P, Nicol P, et al. The duration of oral anticoagulant therapy after a second episode of venous thromboembolism. The Duration of Anticoagulation Trial Study Group. N Engl J Med. 1997;336:393-8. [PMID: 9010144] 8. Kearon C, Gent M, Hirsh J, Weitz J, Kovacs MJ, Anderson DR, et al. A comparison of three months of anticoagulation with extended anticoagulation for a first episode of idiopathic venous thromboembolism. N Engl J Med. 1999;340: 901-7. [PMID: 10089183]. 9. Agnelli G, Prandoni P, Santamaria MG, Bagatella P, Iorio A, Bazzan M, et al. Three months versus one year of oral anticoagulant therapy for idiopathic deep venous thrombosis. Warfarin Optimal Duration Italian Trial Investigators. N Engl J Med. 2001;345:165-9. [PMID: 11463010] 10. Bauer KA. The thrombophilias: well-defined risk factors with uncertain therapeutic implications. Ann Intern Med. 2001;135:367-73. [PMID: 11529700] 11. van den Belt AG, Hutten BA, Prins MH, Bossuy PM. Duration of oral anticoagulant treatment in patients with venous thromboembolism and a deficiency of antithrombin, protein C or protein S—a decision analysis. Thromb Haemost. 2000;84:758-63. [PMID: 11127851] 12. Douketis JD, Foster GA, Crowther MA, Prins MH, Ginsberg JS. Clinical risk factors and timing of recurrent venous thromboembolism during the initial 3 months of anticoagulant therapy. Arch Intern Med. 2000;160:3431-6. [PMID: 11112236] 13. Silverstein MD, Heit JA, Mohr DN, Petterson TM, O’Fallon WM, Melton LJ 3rd. Trends in the incidence of deep vein thrombosis and pulmonary embolism: a 25-year population-based study. Arch Intern Med. 1998;158:58593. [PMID: 9521222] 14. Oger E. Incidence of venous thromboembolism: a community-based study in Western France. EPI-GETBP Study Group. Groupe d’Etude de la Thrombose de Bretagne Occidentale. Thromb Haemost. 2000;83:657-60. [PMID: 10823257] 15. Tsai AW, Cushman M, Rosamond WD, Heckbert SR, Polak JF, Folsom AR. Cardiovascular risk factors and venous thromboembolism incidence: the longitudinal investigation of thromboembolism etiology. Arch Intern Med. 2002; 162:1182-9. [PMID: 12020191] 16. Ridker PM, Goldhaber SZ, Danielson E, Rosenberg Y, Eby CS, Deitcher SR, et al. Long-term, low-intensity warfarin therapy for the prevention of recurrent venous thromboembolism. N Engl J Med. 2003;348:1425-34. [PMID: 12601075] 17. Agnelli G, Prandoni P, Becattini C, Silingardi M, Taliani MR, Miccio M, et al. Extended oral anticoagulant therapy after a first episode of pulmonary www.annals.org

Clinical Impact of Bleeding during Oral Anticoagulant Therapy embolism. Ann Intern Med. 2003;139:19-25. [PMID: 12834314] 18. Landefeld CS, Goldman L. Major bleeding in outpatients treated with warfarin: incidence and prediction by factors known at the start of outpatient therapy. Am J Med. 1989;87:144-52. [PMID: 2787958] 19. Landefeld CS, Beyth RJ. Anticoagulant-related bleeding: clinical epidemiology, prediction, and prevention. Am J Med. 1993;95:315-28. [PMID: 8368229] 20. Anand SS, Yusuf S. Oral anticoagulant therapy in patients with coronary artery disease: a meta-analysis. JAMA. 1999;282:2058-67. [PMID: 10591389] 21. Cappelleri JC, Fiore LD, Brophy MT, Deykin D, Lau J. Efficacy and safety of combined anticoagulant and antiplatelet therapy versus anticoagulant monotherapy after mechanical heart-valve replacement: a metaanalysis. Am Heart J. 1995;130:547-52. [PMID: 7661074] 22. Fihn SD, Callahan CM, Martin DC, McDonell MB, Henikoff JG, White RH. The risk for and severity of bleeding complications in elderly patients treated with warfarin. The National Consortium of Anticoagulation Clinics. Ann Intern Med. 1996;124:970-9. [PMID: 8624064] 23. Hart RG, Benavente O, Pearce LA. Increased risk of intracranial hemorrhage when aspirin is combined with warfarin: a meta-analysis and hypothesis. Cerebrovasc Dis. 1999;9:215-7. [PMID: 10393408] 24. Punthakee X, Doobay J, Anand SS. Oral-anticoagulant-related intracerebral hemorrhage. Thromb Res. 2002;108:31-6. [PMID: 12586129] 25. Pinede L, Duhaut P, Cucherat M, Ninet J, Pasquier J, Boissel JP. Comparison of long versus short duration of anticoagulant therapy after a first episode of venous thromboembolism: a meta-analysis of randomized, controlled trials. J Intern Med. 2000;247:553-62. [PMID: 10809994] 26. Gitter MJ, Jaeger TM, Petterson TM, Gersh BJ, Silverstein MD. Bleeding and thromboembolism during anticoagulant therapy: a population-based study in Rochester, Minnesota. Mayo Clin Proc. 1995;70:725-33. [PMID: 7630209] 27. Hylek EM, Chang YC, Skates SJ, Hughes RA, Singer DE. Prospective study of the outcomes of ambulatory patients with excessive warfarin anticoagulation. Arch Intern Med. 2000;160:1612-7. [PMID: 10847254] 28. College of American Pathologists Coagulation Survey Set 1997 CG2-C. Northfield, IL: College of American Pathologists; 1997. 29. McManus RJ, Wilson S, Delaney BC, Fitzmaurice DA, Hyde CJ, Tobias RS, et al. Review of the usefulness of contacting other experts when conducting a literature search for systematic reviews. BMJ. 1998;317:1562-3. [PMID: 9836655] 30. Direct thrombin inhibitors in acute coronary syndromes: principal results of a meta-analysis based on individual patients’ data. Lancet. 2002;359:294-302. [PMID: 11830196] 31. Fleiss JL. Statistical Methods for Rates and Proportions. New York: Wiley; 1973;:43-7. 32. Levine MN, Hirsh J, Landefeld S, Raskob G. Hemorrhagic complications of anticoagulant treatment. Chest. 1992;102:352S-363S. [PMID: 1395820] 33. Laird NM, Mosteller F. Some statistical methods for combining experimental results. Int J Technol Assess Health Care. 1990;6:5-30. [PMID: 2361819] 34. Low-molecular-weight heparin in the treatment of patients with venous thromboembolism. The Columbus Investigators. N Engl J Med. 1997;337:65762. [PMID: 9280815] 35. Boccalon H, Elias A, Chale JJ, Cadene A, Gabriel S. Clinical outcome and cost of hospital vs home treatment of proximal deep vein thrombosis with a low-molecular-weight heparin: the Vascular Midi-Pyrenees study. Arch Intern Med. 2000;160:1769-73. [PMID: 10871969] 36. Bounameaux H, Ehringer H, Hulting J, Rasche H, Rapold HJ, Zultak M. An exploratory trial of two dosages of a novel synthetic thrombin inhibitor (napsagatran, Ro 46-6240) compared with unfractionated heparin for treatment of proximal deep-vein thrombosis—results of the European multicenter ADVENT trial. Thromb Haemost. 1997;78:997-1002. [PMID: 9308743] 37. Brandjes DP, Heijboer H, Buller HR, de Rijk M, Jagt H, ten Cate JW. Acenocoumarol and heparin compared with acenocoumarol alone in the initial treatment of proximal-vein thrombosis. N Engl J Med. 1992;327:1485-9. [PMID: 1406880] 38. Charbonnier BA, Fiessinger JN, Banga JD, Wenzel E, d’Azemar P, Sagnard L. Comparison of a once daily with a twice daily subcutaneous low molecular weight heparin regimen in the treatment of deep vein thrombosis. FRAXODI group. Thromb Haemost. 1998;79:897-901. [PMID: 9609216] 39. Das SK, Cohen AT, Edmondson RA, Melissari E, Kakkar VV. Low-mowww.annals.org

Article

lecular-weight heparin versus warfarin for prevention of recurrent venous thromboembolism: a randomized trial. World J Surg. 1996;20:521-6; discussion 526-7. [PMID: 8661630] 40. Decousus H, Leizorovicz A, Parent F, Page Y, Tardy B, Girard P, et al. A clinical trial of vena caval filters in the prevention of pulmonary embolism in patients with proximal deep-vein thrombosis. Prevention du Risque d’Embolie Pulmonaire par Interruption Cave Study Group. N Engl J Med. 1998;338:40915. [PMID: 9459643] 41. Gonzalez-Fajardo JA, Arreba E, Castrodeza J, Perez JL, Fernandez L, Agundez I, et al. Venographic comparison of subcutaneous low-molecular weight heparin with oral anticoagulant therapy in the long-term treatment of deep venous thrombosis. J Vasc Surg. 1999;30:283-92. [PMID: 10436448] 42. Harenberg J, Schmidt JA, Koppenhagen K, Tolle A, Huisman MV, Buller HR. Fixed-dose, body weight-independent subcutaneous LMW heparin versus adjusted dose unfractionated intravenous heparin in the initial treatment of proximal venous thrombosis. EASTERN Investigators. Thromb Haemost. 2000;83: 652-6. [PMID: 10823256] 43. Hull RD, Raskob GE, Brant RF, Pineo GF, Elliott G, Stein PD, et al. Low-molecular-weight heparin vs heparin in the treatment of patients with pulmonary embolism. American-Canadian Thrombosis Study Group. Arch Intern Med. 2000;160:229-36. [PMID: 10647762] 44. Hull RD, Raskob GE, Rosenbloom D, Panju AA, Brill-Edwards P, Ginsberg JS, et al. Heparin for 5 days as compared with 10 days in the initial treatment of proximal venous thrombosis. N Engl J Med. 1990;322:1260-4. [PMID: 2183055] 45. Kovacs MJ, Cruickshank M, Wells PS, Kim H, Chin-Yee I, Morrow B, et al. Randomized assessment of a warfarin nomogram for initial oral anticoagulation after venous thromboembolic disease. Haemostasis. 1998;28:62-9. [PMID: 10087430] 46. Koopman MM, Prandoni P, Piovella F, Ockelford PA, Brandjes DP, van der Meer J, et al. Treatment of venous thrombosis with intravenous unfractionated heparin administered in the hospital as compared with subcutaneous lowmolecular-weight heparin administered at home. The Tasman Study Group. N Engl J Med. 1996;334:682-7. [PMID: 8594426] 47. Leroyer C, Bressollette L, Oger E, Mansourati J, Cheze-Le Rest C, Nonent M, et al. Early versus delayed introduction of oral vitamin K antagonists in combination with low-molecular-weight heparin in the treatment of deep vein thrombosis. a randomized clinical trial. The ANTENOX Study Group. Haemostasis. 1998;28:70-7. [PMID: 10087431] 48. Levine MN, Hirsh J, Gent M, Turpie AG, Weitz J, Ginsberg J, et al. Optimal duration of oral anticoagulant therapy: a randomized trial comparing four weeks with three months of warfarin in patients with proximal deep vein thrombosis. Thromb Haemost. 1995;74:606-11. [PMID: 8584992] 49. Lopaciuk S, Meissner AJ, Filipecki S, Zawilska K, Sowier J, Ciesielski L, et al. Subcutaneous low molecular weight heparin versus subcutaneous unfractionated heparin in the treatment of deep vein thrombosis: a Polish multicenter trial. Thromb Haemost. 1992;68:14-8. [PMID: 1325076] 50. Lopaciuk S, Bielska-Falda H, Noszczyk W, Bielawiec M, Witkiewicz W, Filipecki S, et al. Low molecular weight heparin versus acenocoumarol in the secondary prophylaxis of deep vein thrombosis. Thromb Haemost. 1999;81:2631. [PMID: 9974369] 51. Merli G, Spiro TE, Olsson CG, Abildgaard U, Davidson BL, Eldor A, et al. Subcutaneous enoxaparin once or twice daily compared with intravenous unfractionated heparin for treatment of venous thromboembolic disease. Ann Intern Med. 2001;134:191-202. [PMID: 11177331] 52. Pinede L, Ninet J, Duhaut P, Chabaud S, Demolombe-Rague S, Durieu I, et al. Comparison of 3 and 6 months of oral anticoagulant therapy after a first episode of proximal deep vein thrombosis or pulmonary embolism and comparison of 6 and 12 weeks of therapy after isolated calf deep vein thrombosis. Circulation. 2001;103:2453-60. [PMID: 11369685] 53. Prandoni P, Lensing AW, Buller HR, Carta M, Cogo A, Vigo M, et al. Comparison of subcutaneous low-molecular-weight heparin with intravenous standard heparin in proximal deep-vein thrombosis. Lancet. 1992;339:441-5. [PMID: 1346817] 54. Simonneau G, Sors H, Charbonnier B, Page Y, Laaban JP, Azarian R, et al. A comparison of low-molecular-weight heparin with unfractionated heparin for acute pulmonary embolism. The THESEE Study Group. Tinzaparine ou Heparine Standard: Evaluations dans l’Embolie Pulmonaire. N Engl J Med. 1997; 2 December 2003 Annals of Internal Medicine Volume 139 • Number 11 899

Article

Clinical Impact of Bleeding during Oral Anticoagulant Therapy

337:663-9. [PMID: 9278462] 55. Treatment of proximal deep vein thrombosis with a novel synthetic compound (SR90107A/ORG31540) with pure anti-factor Xa activity: a phase II evaluation. The Rembrandt Investigators. Circulation. 2000;102:2726-31. [PMID: 11094039] 56. Breddin HK, Hach-Wunderle V, Nakov R, Kakkar VV. Effects of a lowmolecular-weight heparin on thrombus regression and recurrent thromboembolism in patients with deep-vein thrombosis. N Engl J Med. 2001;344:626-31. [PMID: 11228276] 57. Hull RD, Raskob GE, Pineo GF, Green D, Trowbridge AA, Elliott CG, et al. Subcutaneous low-molecular-weight heparin compared with continuous intravenous heparin in the treatment of proximal-vein thrombosis. N Engl J Med. 1992;326:975-82. [PMID: 1545850] 58. Pini M, Aiello S, Manotti C, Pattacini C, Quintavalla R, Poli T, et al. Low molecular weight heparin versus warfarin in the prevention of recurrences after deep vein thrombosis. Thromb Haemost. 1994;72:191-7. [PMID: 7831650] 59. Lindmarker P, Holmstrom M. Use of low molecular weight heparin (dalteparin), once daily, for the treatment of deep vein thrombosis. A feasibility and health economic study in an outpatient setting. Swedish Venous Thrombosis Dalteparin Trial Group. J Intern Med. 1996;240:395-401. [PMID: 9010387] 60. Monreal M, Roncales FJ, Ruiz J, Muchart J, Fraile M, Costa J, et al. Secondary prevention of venous thromboembolism: a role for low-molecularweight heparin. Haemostasis. 1998;28:236-43. [PMID: 10420072] 61. Wells PS, Kovacs MJ, Bormanis J, Forgie MA, Goudie D, Morrow B, et al. Expanding eligibility for outpatient treatment of deep venous thrombosis and

900 2 December 2003 Annals of Internal Medicine Volume 139 • Number 11

pulmonary embolism with low-molecular-weight heparin: a comparison of patient self-injection with homecare injection. Arch Intern Med. 1998;158:180912. [PMID: 9738611] 62. Kovacs MJ, Anderson D, Morrow B, Gray L, Touchie D, Wells PS. Outpatient treatment of pulmonary embolism with dalteparin. Thromb Haemost. 2000;83:209-11. [PMID: 10739374] 63. Harrison L, Johnston M, Massicotte MP, Crowther M, Moffat K, Hirsh J. Comparison of 5-mg and 10-mg loading doses in initiation of warfarin therapy. Ann Intern Med. 1997;126:133-6. [PMID: 9005747] 64. Kovacs MJ, Rodger M, Anderson DR, Morrow B, Kells G, Kovacs J, et al. Comparison of 10-mg and 5-mg warfarin initiation nomograms together with low-molecular-weight heparin for outpatient treatment of acute venous thromboembolism. A randomized, double-blind, controlled trial. Ann Intern Med. 2003;138:714-9. [PMID: 12729425] 65. Levine MN, Raskob G, Hirsh J. Hemorrhagic complications of long-term anticoagulant therapy. Chest. 1986;89:16S-25S. [PMID: 3510826] 66. Levine MN, Raskob G, Landefeld S, Kearon C. Hemorrhagic complications of anticoagulant treatment. Chest. 2001;119:108S-121S. [PMID: 11157645] 67. Palareti G, Leali N, Coccheri S, Poggi M, Manotti C, D’Angelo A, et al. Bleeding complications of oral anticoagulant treatment: an inception-cohort, prospective collaborative study (ISCOAT). Italian Study on Complications of Oral Anticoagulant Therapy. Lancet. 1996;348:423-8. [PMID: 8709780] 68. Hylek EM, Singer DE. Risk factors for intracranial hemorrhage in outpatients taking warfarin. Ann Intern Med. 1994;120:897-902. [PMID: 8172435]

www.annals.org

Current Author Addresses: Dr. Linkins: Henderson Research Centre,

Room 220, 711 Concession Street, Hamilton, Ontario L8V 1C3, Canada. Dr. Choi: Department of Anesthesia, University of British Columbia, 2329 West Mall, Vancouver V6T 1Z4, British Columbia. Dr. Douketis: St. Joseph’s Healthcare, Room F-541, 50 Charlton Avenue East, Hamilton, Ontario L8N 4A6, Canada. Author Contributions: Conception and design: L.A. Linkins, J.D.

Douketis.

www.annals.org

Analysis and interpretation of the data: L.A. Linkins, P.T. Choi, J.D. Douketis. Drafting of the article: L.A. Linkins, J.D. Douketis. Critical revision of the article for important intellectual content: L.A. Linkins, P.T. Choi, J.D. Douketis. Final approval of the article: L.A. Linkins, P.T. Choi, J.D. Douketis. Statistical expertise: P.T. Choi. Administrative, technical, or logistic support: J.D. Douketis. Collection and assembly of data: L.A. Linkins.

Annals of Internal Medicine Volume • Number

E-901