Stroke and recurrent haemorrhage associated with antithrombotic treatment after gastrointestinal bleeding in patients with atrial fibrillation: nationwide cohort study
BMJ 2015; 351 doi: https://doi.org/10.1136/bmj.h5876 (Published 16 November 2015) Cite this as: BMJ 2015;351:h5876
All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
Dear Dr. Beales
Thank you for your interest in our study and for the discussion points regarding the methods and the results.
As to the first comment, it is correct that today’s guidelines for atrial fibrillation do not include antiplatelet agents for stroke prophylaxis. However, we found it was important to examine all possible antithrombotic treatment regimens after a gastrointestinal bleeding to compare the effectiveness and safety of oral anticoagulation versus non-resumption or antiplatelet agents, and the results suggested that independent of antithrombotic treatment regimen before the gastrointestinal bleeding, patients with atrial fibrillation benefited most from restarting or modifying treatment to oral anticoagulants after a gastrointestinal bleed.
One of the main messages in our study was that after a gastrointestinal bleeding it is important to consider restart of oral anticoagulation, but the management of these high risk patients, including patients with the appearance of ‘non-H pylori, non-drug' (NSAIDs/aspirin) associated ulcers, is a challenging task. We agree that the patient’s profile/individualised factors and shared decision making also should be taken into account when deciding whether to restart or withhold anticoagulant treatment after a gastrointestinal bleeding.
In a clinical everyday situation, it would be reasonable to consider restarting oral anticoagulation before 90 days after surviving a gastrointestinal bleeding. In our observational study, we chose a blanking period of 90 days since the numbers of tablets in the largest package size available for antithrombotics last for approximately three months. We were then able to eliminate misclassification—for instance, after discharge a patient could initially use antithrombotics from an old package from before the gastrointestinal bleeding, and because no new prescription would had been dispensed immediately after discharge, this patient would be registered as non-resumption if we did not include the 90 days blanking period, despite the patient actually having restarted antithrombotic treatment. With the 90 days blanking period, we allowed for a new prescription of antithrombotics to be dispensed for all the patients. We discussed the 90 days blanking period in the discussion section and added extra sensitivity analyses with a 30, 60, and 120 days blanking period. Overall, the results were similar to the main results. To evaluate when decisions to restart oral anticoagulation should be made would be a relevant future research project, because our observational study was limited to give an answer to this question.
We thank Dr. Beales for contributing with relevant comments on our research paper.
Competing interests: No competing interests
Dear Editors
It is a curious thing that some researchers lauded the apparent better effectiveness and safety outcome of AF patients after restarting oral anticoagulation as anti-thrombotic therapy on 90-day survivors of GI bleed; the cohort under analysis excluded patients who died from complications from GI bleed (in which the antithrombotic therapy itself invariably significantly contributes to the causation of the bleed) as well as failing to adjust for those who are not restarted on therapy after being judged as too frail to restart anti-thrombotic treatment (which the comorbidities contributing to frailty wlll increase mortality independently).
A stunning endorsement of safety and effectiveness of oral anticoagulant? I think not.
Competing interests: No competing interests
The paper provides support for the important, but perhaps underused, practice of reintroducing anticoagulation after a gastrointestinal haemorrhage. Some aspects of the paper should be placed in context. The data on anti-platelet agents for atrial fibrillation probably have little relevance to current practice. Over the time course of the study (1996-2012) the management and epidemiology of gastrointestinal bleeding have changed which could have important impacts on the conclusions for practice today. Improvements in endoscopic and medical management have reduced early and late rebleeding rates but possibly more importantly the causes of acute upper GI bleeding have changed, with a marked reduction in H pylori-associated ulcers and the appearance of ‘non-H pylori, non-drug' (NSAIDs/aspirin) associated ulcers (1). This latter group is important because they are associated with a very high risk of rebleeding that is not ameliorated by acid suppressive drugs as well as a high rate of subsequent morbidity (1,2). The risk/benefit equation of anticoagulation in these patients is uncertain and reinforces that such decisions must be based on more individualised factors than outlined in the study.
In addition, the inference, secondary to the methodology employed, is that decisions about resuming anticoagulation should be made 3 months after a GI bleed. This is almost certainly too late. The authors draw parallels with the anti-platelet literature, where early (even as early as immediately after endoscopic haemostasis of high-risk lesions) drug reintroduction is beneficial; a small increase in early rebleeding is clearly more than offset by improved mortality and circulatory mortality that is seen as early as 30 days and continues for 6 months (3,4). Early anticoagulation reintroduction in this situation is also beneficial. The peak time for rebleeding is 13 day post-discharge and most cases of thromboembolism occur 2-8 weeks after the initial GI bleeding event (5). Patients discharged from hospital on anticoagulation had significantly lower rates of thromboembolism (hazard ratio 0.121, 95% confidence intervals 0.006-0.812) without significantly increased rebleeding rates compared to not restarting anticoagulation at discharge (5). Similar results have been reported in those with reintroduction within 30 days compared to later (6). Therefore it is important that decisions on anticoagulation reintroduction are made earlier but the underlying pathology must be considered.
References
1. Prevention of upper gastrointestinal haemorrhage: current controversies and clinical guidance. Brooks J, Warburton R, Beales IL. Ther Adv Chronic Dis. 2013; 4(5):206-22
2. Gastroprotective therapy does not improve outcomes of patients with Helicobacter pylori-negative idiopathic bleeding ulcers. Wong GL, Au KW, Lo AO, Tse YK, Ching JY, To KF, Chan FK. Clin Gastroenterol Hepatol. 2012 10(10):1124-9.
3. Continuation of low-dose aspirin therapy in peptic ulcer bleeding: a randomized trial. Sung JJ, Lau JY, Ching JY, Wu JC, Lee YT, Chiu PW, Leung VK, Wong VW, Chan FK. Ann Intern Med. 2010;152(1):1-9
4. Discontinuation of Low-Dose Aspirin Therapy After Peptic Ulcer Bleeding Increases Risk of Death and Acute Cardiovascular Events. Derogar M, Sandblom G, Lundell L, Orsini N, Bottai M, Lu Y, Sadr–Azodi O. Clinical Gastroenterology and Hepatology, 2013 11 (1) 38-42
5. The risks of thromboembolism vs. recurrent gastrointestinal bleeding after interruption of systemic anticoagulation in hospitalized inpatients with gastrointestinal bleeding: a prospective study. Sengupta N, Feuerstein JD, Patwardhan VR, Tapper EB, Ketwaroo GA, Thaker AM, Leffler DA. Am J Gastroenterol. 2015:110(2):328-35.
6. Restarting anticoagulation and outcomes after major gastrointestinal bleeding in atrial fibrillation. Alirhayim Z, Khalid F, Qureshi W. Am J Cardiol 2014; 113: 662-668
Competing interests: No competing interests
I really do not understand the rapid response to the paper. The authors specifically say "owing to the paucity of data on non-vitamin K antagonists oral anticoagulants in the current study we cannot conclude anything on the efficacy or safety of these drugs." This is the correct interpretation, it is simply impossible to draw any conclusions about these newer drugs.
Competing interests: No competing interests
Sir,
Patients with atrial fibrillation are at higher risk of arterial thromboembolism & intracranial haemorrhage and therefore require antithrombotic treatment. However, if such patients have an episode of gastrointestinal bleed as a side effect of antithrombotic treatment, the decision to resume the antithrombotic treatment is often difficult to make due to almost an equal risk benefit ratio. In this context, this study is one of the kind to clearly “highlight” the importance of resuming the antithrombotic therapy with the newer oral anticoagulant drugs even after gastrointestinal bleed has occurred because the benefits clearly outweigh the risks associated and lead to reduction of mortality due to thrombotic complications.
The newer oral anticoagulants have been stated to be equally effective as age old warfarin therapy, if not more, in the studies done for the same, and they hadve the added advantages of increased patient compliance because of predictable pharmacokinetics and lesser drug interactions. The associated bleeding risk also with these drugs is almost equal to that of warfarin.
One of the meta analysis of the drugs rivaroxaban, dabigatran and apixaban versus warfarin showed that these newer oral anticoagulants have a higher efficacy compared to warfarin when it is about prevention of stroke and systemic embolism in patients with AF. Additionally, it was seen that they were associated with lower risk for intracranial bleeding. Hence, tipping the overall risk benefit ratio to “favourable”1. In another similar study, it was again seen that these newer oral anticoagulants have an edge of providing better risk benefit ratio over warfarin in patients with atrial fibrillation.2
This study by Staerk L et al clearly showed the beneficial risk benefit ratio of these newer drugs, over both warfarin and antiplatelet agents, thereby putting these newer anticoagulants on a safer pedestal than before. In this study, it was seen that newer anticoagulants when given alone as compared to the combination antithrombotic therapy had better effects in the patients with atrial fibrillation and gastrointestinal bleed, as benefits were more than the risk of re bleeding. The newer anticoagulants fared better than the antiplatelet agents and warfarin and are slowly coming out stronger especially in high risk patients.
References:
1. Miller CS, Grandi SM, Shimony A, Filion KB, Eisenberg MJ. Meta-analysis of efficacy and safety of new oral anticoagulants (dabigatran, rivaroxaban, apixaban) versus warfarin in patients with atrial fibrillation. Am J Cardiol. 2012;110(3):453-60.
2. Harenberg J, Marx S, Diener HC, Lip GY, Marder VJ, Wehling M et al. Comparison of efficacy and safety of dabigatran, rivaroxaban and apixaban in patients with atrial fibrillation using network meta-analysis. Int Angiol. 2012;31(4):330-9.
Competing interests: No competing interests
Re: A matter of perspective, or the lack of it
Dear Dr. Goh
Thank you for your comment.
One of our research questions was whether restarting oral anticoagulation in ‘survivors’ of gastrointestinal bleed would do better in the long run compared with non-resumption. Some clinicians may just stop antithrombotic drugs forever after a gastrointestinal bleeding, but our results suggested that it is important to consider restart of oral anticoagulation after a gastrointestinal bleeding (restart of oral anticoagulation was associated with better outcomes for all cause mortality and thromboembolism, but an increased longitudinal associated risk of bleeding). There is no easy way for frailty assessment in registries, but we have adjusted for age and multiple comorbidities in the statistical analyses.
We thank Dr. Goh for contributing with relevant comments on our research paper.
Competing interests: No competing interests