Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You and Your Team.

Learn More →

Resuming Anticoagulation in the First Week Following Gastrointestinal Tract Hemorrhage: Comment on “Risk of Thromboembolism, Recurrent Hemorrhage, and Death After Warfarin Therapy Interruption for Gastrointestinal Tract Bleeding”

Resuming Anticoagulation in the First Week Following Gastrointestinal Tract Hemorrhage: Comment... Gastrointestinal tract (GI) bleeding is one of the most frequently encountered, and feared, complications of warfarin therapy. However, we know surprisingly little about whether or when to resume warfarin therapy following GI hemorrhage. Through the lens of “first do no harm,” it may seem risky—if not foolish—to resume a drug therapy that recently contributed to a life-threatening medical complication, let alone doing so within the first week following the bleeding event. Yet because warfarin is used to prevent devastating and potentially fatal thrombotic events, clinicians may feel that they have no choice but to resume treatment and hope for the best. Furthermore, some thrombotic events, such as pulmonary embolism and atrial fibrillation–related stroke, are far more likely to occur during brief periods of warfarin therapy cessation than would be expected, based on extrapolation from annualized event rates in nonanticoagulated patients, particularly in the setting of invasive procedures.1,2 This suggests that warfarin therapy should be resumed as soon as possible following most anticoagulant interruptions. But how soon is too soon to reanticoagulate following GI hemorrhage? The recent American College of Chest Physicians (ACCP) guidelines on anticoagulation management offer strategies for handling the acute phase of anticoagulant-associated hemorrhage using 4-factor prothrombin complex if available or fresh-frozen plasma if not, combined with intravenous phytonadione.3 However, these guidelines offer no guidance on the timing of anticoagulant reinitiation. There are small case series suggesting that the risk for both rebleeding (when anticoagulation is resumed) and thrombosis (if it is not resumed) may be substantial in the 1 to 3 months following warfarin-associated GI bleeding,4,5 but data are too sparse to offer definitive guidance. In this issue of the Archives, Witt and colleagues6 present important findings on the risks and benefits of resuming anticoagulation in patients with warfarin-associated GI bleeding. Harnessing both administrative and clinical data from Kaiser Permanente Colorado's Clinical Pharmacy Anticoagulation Service, they identified 442 patients with confirmed warfarin-associated GI bleeding and characterized the clinical presentation of the initial bleeding episode and its management in the acute phase, the timing of warfarin therapy reinitiation (if done), and the ultimate clinical outcomes. Warfarin therapy was resumed in 260 patients after a median of 4 days (interquartile range, 2-9 days) and was not resumed in the remaining 182 patients. Interestingly, almost half of the patients included in this cohort also had been taking aspirin at the time of the bleeding. Not unexpectedly, patients in whom warfarin therapy was resumed tended to be younger than those in whom it was withheld, were more likely to have prosthetic heart valves, and were less likely to have had an index bleeding event that was serious enough to require blood products or intensive care unit admission. After rigorous adjustment for these important differences using propensity score methods, patients who resumed warfarin therapy had a 3-fold lower risk of death, a more than 10-fold reduction in thrombotic events, and only a modest and nonsignificant increased risk of recurrent GI bleeding than those in whom warfarin was withheld. Overall, 12% of the patients in this cohort were dead within 3 months, including more than 20% of those who did not resume warfarin therapy and 6% of those who did. The majority of thrombotic events and deaths occurred within 30 days from the index GI hemorrhage, but, importantly, the majority of deaths were not attributed to thrombosis. In particular, malignancy accounted for more than a quarter of deaths (29%), with infection and cardiac disease accounting for a large percentage of the remainder. The authors were careful to exclude from their analysis those deaths that occurred within the first week from the index hemorrhage, since these deaths were likely related to the bleeding episode itself. No subsequent deaths were attributed to recurrent GI bleeding. How can we explain these findings? There are theoretical concerns for rebound hypercoagulability during abrupt warfarin therapy cessation, which might be exacerbated by blood product administration, reduced mobility, and hemodynamic stress occurring in the setting of GI hemorrhage. However, the low incidence of thrombotic deaths observed in this study suggests that there may have been unmeasured clinical factors (or patient preferences) that accounted for much of the apparent benefit afforded by resuming anticoagulation. Some patients, for instance, may have had GI hemorrhage as the presenting symptom of advanced malignancy and may have declined to resume anticoagulation on this basis. Indeed, residual confounding is likely present in any observational study that involves allocation to potentially risky therapies such as anticoagulation, regardless of the statistical rigor, but it is also plausible that subclinical thrombosis contributed to some of the excess deaths, particularly those attributed to cardiovascular causes. The uncertainty surrounding the causal effect of anticoagulation strategy on observed death rates does not detract from the study's key findings. First, this study demonstrates that physicians and patients are willing, in most instances, to resume anticoagulation after GI bleeding. Second, anticoagulation was generally resumed within the first week of the event (a median of 4 days after hemorrhage). Third, among the 260 patients who resumed anticoagulation, the risk of recurrent bleeding was acceptably low (10%) and there were no fatal recurrent GI bleeding episodes. On the basis of these observations and in the absence of other studies providing competing data, we believe that most patients with warfarin-associated GI bleeding and indications for continued long-term antithrombotic therapy should resume anticoagulation within the first week following the hemorrhage, approximately 4 days afterward, if we use the median anticoagulation reinitiation time in this study as a benchmark. Although not specifically addressed in this study, we would hesitate to continue concurrent antiplatelet therapy in these patients without a compelling indication to do so (such as a recent coronary stent)7 and also would caution against extrapolating these findings to newer anticoagulants, such as dabigatran and rivaroxaban, that may be associated with more GI bleeding than warfarin when used long term8,9 and whose effects are not easily reversed. Finally, while it is tempting to advocate for prospective, randomized trials to address the long-term management of anticoagulant-associated GI bleeding, we doubt that any such trials are logistically feasible or palatable to health care providers and patients. As such, high-quality, real-world data, such as those presented by Witt and colleagues,6 can—and should—form the foundation of our evidence base. Back to top Article Information Correspondence: Dr Brotman, Hospitalist Program, Johns Hopkins Hospital, 600 N Wolfe St, Park 307, Baltimore, MD 21287 (brotman@jhmi.edu). Published Online: September 17, 2012. doi:10.1001/archinternmed.2012.4309 Financial Disclosure: Dr Brotman has served on advisory boards or as a consultant for Gerson Lehrman Group (miscellaneous consulting activities—current), The Dunn Group (miscellaneous consulting activities—current), and Quantia Communications, LLC (hospitalist leadership panel—current), and received research support from Siemens Healthcare Diagnostics (formerly Dade Behring) (project completed in 2010), Agency for Healthcare Research and Quality, Centers for Medicare & Medicaid Services, and Amerigroup Corporation. Dr Jaffer has served as a consultant for sanofi-aventis, Boehringer Ingelheim, Bristol Myers Squibb, Janssen Pharmaceuticals, Canyon Pharmaceuticals, and CSL Behring and received research and grant support from the National Heart, Lung, and Blood Institute. References 1. Jaffer AK, Brotman DJ, Bash LD, Mahmood SK, Lott B, White RH. Variations in perioperative warfarin management: outcomes and practice patterns at nine hospitals. Am J Med. 2010;123(2):141-15020103023PubMedGoogle ScholarCrossref 2. Kaatz S, Douketis JD, Zhou H, Gage BF, White RH. Risk of stroke after surgery in patients with and without chronic atrial fibrillation. J Thromb Haemost. 2010;8(5):884-89020096001PubMedGoogle Scholar 3. Holbrook A, Schulman S, Witt DM, et al; American College of Chest Physicians. Evidence-based management of anticoagulant therapy: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest. 2012;141(2):(suppl) e152S-e184S22315259PubMedGoogle ScholarCrossref 4. Lee JK, Kang HW, Kim SG, Kim JS, Jung HC. Risks related with withholding and resuming anticoagulation in patients with non-variceal upper gastrointestinal bleeding while on warfarin therapy. Int J Clin Pract. 2012;66(1):64-6822171905PubMedGoogle ScholarCrossref 5. Ananthasubramaniam K, Beattie JN, Rosman HS, Jayam V, Borzak S. How safely and for how long can warfarin therapy be withheld in prosthetic heart valve patients hospitalized with a major hemorrhage? Chest. 2001;119(2):478-48411171726PubMedGoogle ScholarCrossref 6. Witt DM, Delate T, Garcia DA, et al. Risk of thromboembolism, recurrent hemorrhage, and death after warfarin therapy interruption for gastrointestinal tract bleeding [published online September 17, 2012]. Arch Intern Med. 2012;172(19):ioi1200541484-1491Google Scholar 7. Hansen ML, Sørensen R, Clausen MT, et al. Risk of bleeding with single, dual, or triple therapy with warfarin, aspirin, and clopidogrel in patients with atrial fibrillation. Arch Intern Med. 2010;170(16):1433-144120837828PubMedGoogle ScholarCrossref 8. Connolly SJ, Ezekowitz MD, Yusuf S, et al; RE-LY Steering Committee and Investigators. Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med. 2009;361(12):1139-115119717844PubMedGoogle ScholarCrossref 9. Patel MR, Mahaffey KW, Garg J, et al; ROCKET AF Investigators. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med. 2011;365(10):883-89121830957PubMedGoogle ScholarCrossref http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Archives of Internal Medicine American Medical Association

Resuming Anticoagulation in the First Week Following Gastrointestinal Tract Hemorrhage: Comment on “Risk of Thromboembolism, Recurrent Hemorrhage, and Death After Warfarin Therapy Interruption for Gastrointestinal Tract Bleeding”

Loading next page...
 
/lp/american-medical-association/resuming-anticoagulation-in-the-first-week-following-gastrointestinal-3H8oOkSYJW
Publisher
American Medical Association
Copyright
Copyright © 2012 American Medical Association. All Rights Reserved.
ISSN
0003-9926
eISSN
1538-3679
DOI
10.1001/archinternmed.2012.4309
Publisher site
See Article on Publisher Site

Abstract

Gastrointestinal tract (GI) bleeding is one of the most frequently encountered, and feared, complications of warfarin therapy. However, we know surprisingly little about whether or when to resume warfarin therapy following GI hemorrhage. Through the lens of “first do no harm,” it may seem risky—if not foolish—to resume a drug therapy that recently contributed to a life-threatening medical complication, let alone doing so within the first week following the bleeding event. Yet because warfarin is used to prevent devastating and potentially fatal thrombotic events, clinicians may feel that they have no choice but to resume treatment and hope for the best. Furthermore, some thrombotic events, such as pulmonary embolism and atrial fibrillation–related stroke, are far more likely to occur during brief periods of warfarin therapy cessation than would be expected, based on extrapolation from annualized event rates in nonanticoagulated patients, particularly in the setting of invasive procedures.1,2 This suggests that warfarin therapy should be resumed as soon as possible following most anticoagulant interruptions. But how soon is too soon to reanticoagulate following GI hemorrhage? The recent American College of Chest Physicians (ACCP) guidelines on anticoagulation management offer strategies for handling the acute phase of anticoagulant-associated hemorrhage using 4-factor prothrombin complex if available or fresh-frozen plasma if not, combined with intravenous phytonadione.3 However, these guidelines offer no guidance on the timing of anticoagulant reinitiation. There are small case series suggesting that the risk for both rebleeding (when anticoagulation is resumed) and thrombosis (if it is not resumed) may be substantial in the 1 to 3 months following warfarin-associated GI bleeding,4,5 but data are too sparse to offer definitive guidance. In this issue of the Archives, Witt and colleagues6 present important findings on the risks and benefits of resuming anticoagulation in patients with warfarin-associated GI bleeding. Harnessing both administrative and clinical data from Kaiser Permanente Colorado's Clinical Pharmacy Anticoagulation Service, they identified 442 patients with confirmed warfarin-associated GI bleeding and characterized the clinical presentation of the initial bleeding episode and its management in the acute phase, the timing of warfarin therapy reinitiation (if done), and the ultimate clinical outcomes. Warfarin therapy was resumed in 260 patients after a median of 4 days (interquartile range, 2-9 days) and was not resumed in the remaining 182 patients. Interestingly, almost half of the patients included in this cohort also had been taking aspirin at the time of the bleeding. Not unexpectedly, patients in whom warfarin therapy was resumed tended to be younger than those in whom it was withheld, were more likely to have prosthetic heart valves, and were less likely to have had an index bleeding event that was serious enough to require blood products or intensive care unit admission. After rigorous adjustment for these important differences using propensity score methods, patients who resumed warfarin therapy had a 3-fold lower risk of death, a more than 10-fold reduction in thrombotic events, and only a modest and nonsignificant increased risk of recurrent GI bleeding than those in whom warfarin was withheld. Overall, 12% of the patients in this cohort were dead within 3 months, including more than 20% of those who did not resume warfarin therapy and 6% of those who did. The majority of thrombotic events and deaths occurred within 30 days from the index GI hemorrhage, but, importantly, the majority of deaths were not attributed to thrombosis. In particular, malignancy accounted for more than a quarter of deaths (29%), with infection and cardiac disease accounting for a large percentage of the remainder. The authors were careful to exclude from their analysis those deaths that occurred within the first week from the index hemorrhage, since these deaths were likely related to the bleeding episode itself. No subsequent deaths were attributed to recurrent GI bleeding. How can we explain these findings? There are theoretical concerns for rebound hypercoagulability during abrupt warfarin therapy cessation, which might be exacerbated by blood product administration, reduced mobility, and hemodynamic stress occurring in the setting of GI hemorrhage. However, the low incidence of thrombotic deaths observed in this study suggests that there may have been unmeasured clinical factors (or patient preferences) that accounted for much of the apparent benefit afforded by resuming anticoagulation. Some patients, for instance, may have had GI hemorrhage as the presenting symptom of advanced malignancy and may have declined to resume anticoagulation on this basis. Indeed, residual confounding is likely present in any observational study that involves allocation to potentially risky therapies such as anticoagulation, regardless of the statistical rigor, but it is also plausible that subclinical thrombosis contributed to some of the excess deaths, particularly those attributed to cardiovascular causes. The uncertainty surrounding the causal effect of anticoagulation strategy on observed death rates does not detract from the study's key findings. First, this study demonstrates that physicians and patients are willing, in most instances, to resume anticoagulation after GI bleeding. Second, anticoagulation was generally resumed within the first week of the event (a median of 4 days after hemorrhage). Third, among the 260 patients who resumed anticoagulation, the risk of recurrent bleeding was acceptably low (10%) and there were no fatal recurrent GI bleeding episodes. On the basis of these observations and in the absence of other studies providing competing data, we believe that most patients with warfarin-associated GI bleeding and indications for continued long-term antithrombotic therapy should resume anticoagulation within the first week following the hemorrhage, approximately 4 days afterward, if we use the median anticoagulation reinitiation time in this study as a benchmark. Although not specifically addressed in this study, we would hesitate to continue concurrent antiplatelet therapy in these patients without a compelling indication to do so (such as a recent coronary stent)7 and also would caution against extrapolating these findings to newer anticoagulants, such as dabigatran and rivaroxaban, that may be associated with more GI bleeding than warfarin when used long term8,9 and whose effects are not easily reversed. Finally, while it is tempting to advocate for prospective, randomized trials to address the long-term management of anticoagulant-associated GI bleeding, we doubt that any such trials are logistically feasible or palatable to health care providers and patients. As such, high-quality, real-world data, such as those presented by Witt and colleagues,6 can—and should—form the foundation of our evidence base. Back to top Article Information Correspondence: Dr Brotman, Hospitalist Program, Johns Hopkins Hospital, 600 N Wolfe St, Park 307, Baltimore, MD 21287 (brotman@jhmi.edu). Published Online: September 17, 2012. doi:10.1001/archinternmed.2012.4309 Financial Disclosure: Dr Brotman has served on advisory boards or as a consultant for Gerson Lehrman Group (miscellaneous consulting activities—current), The Dunn Group (miscellaneous consulting activities—current), and Quantia Communications, LLC (hospitalist leadership panel—current), and received research support from Siemens Healthcare Diagnostics (formerly Dade Behring) (project completed in 2010), Agency for Healthcare Research and Quality, Centers for Medicare & Medicaid Services, and Amerigroup Corporation. Dr Jaffer has served as a consultant for sanofi-aventis, Boehringer Ingelheim, Bristol Myers Squibb, Janssen Pharmaceuticals, Canyon Pharmaceuticals, and CSL Behring and received research and grant support from the National Heart, Lung, and Blood Institute. References 1. Jaffer AK, Brotman DJ, Bash LD, Mahmood SK, Lott B, White RH. Variations in perioperative warfarin management: outcomes and practice patterns at nine hospitals. Am J Med. 2010;123(2):141-15020103023PubMedGoogle ScholarCrossref 2. Kaatz S, Douketis JD, Zhou H, Gage BF, White RH. Risk of stroke after surgery in patients with and without chronic atrial fibrillation. J Thromb Haemost. 2010;8(5):884-89020096001PubMedGoogle Scholar 3. Holbrook A, Schulman S, Witt DM, et al; American College of Chest Physicians. Evidence-based management of anticoagulant therapy: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest. 2012;141(2):(suppl) e152S-e184S22315259PubMedGoogle ScholarCrossref 4. Lee JK, Kang HW, Kim SG, Kim JS, Jung HC. Risks related with withholding and resuming anticoagulation in patients with non-variceal upper gastrointestinal bleeding while on warfarin therapy. Int J Clin Pract. 2012;66(1):64-6822171905PubMedGoogle ScholarCrossref 5. Ananthasubramaniam K, Beattie JN, Rosman HS, Jayam V, Borzak S. How safely and for how long can warfarin therapy be withheld in prosthetic heart valve patients hospitalized with a major hemorrhage? Chest. 2001;119(2):478-48411171726PubMedGoogle ScholarCrossref 6. Witt DM, Delate T, Garcia DA, et al. Risk of thromboembolism, recurrent hemorrhage, and death after warfarin therapy interruption for gastrointestinal tract bleeding [published online September 17, 2012]. Arch Intern Med. 2012;172(19):ioi1200541484-1491Google Scholar 7. Hansen ML, Sørensen R, Clausen MT, et al. Risk of bleeding with single, dual, or triple therapy with warfarin, aspirin, and clopidogrel in patients with atrial fibrillation. Arch Intern Med. 2010;170(16):1433-144120837828PubMedGoogle ScholarCrossref 8. Connolly SJ, Ezekowitz MD, Yusuf S, et al; RE-LY Steering Committee and Investigators. Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med. 2009;361(12):1139-115119717844PubMedGoogle ScholarCrossref 9. Patel MR, Mahaffey KW, Garg J, et al; ROCKET AF Investigators. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med. 2011;365(10):883-89121830957PubMedGoogle ScholarCrossref

Journal

Archives of Internal MedicineAmerican Medical Association

Published: Oct 22, 2012

Keywords: anticoagulation,warfarin,thromboembolism,hemorrhage,gastrointestinal bleeding,gastrointestinal tract

References