Anticoagulation with warfarin is the leading cause of emergency hospitalizations resulting from adverse drug events in elderly Americans [1]. In one-third of those events, warfarin alone or in combination with other dugs was identified as the cause. Antiplatelet agents constituted the third leading cause (13.3%), and thus, in half of the cases, antithrombotic therapy was implicated as the cause of emergency hospitalization. Fear of bleeding and old age of the patients are among the most common reasons for not prescribing anticoagulant therapy to patients with atrial fibrillation for whom this kind of stroke prophylaxis is indicated [2]. In a large trial on clopidogrel added to aspirin in patients with atrial fibrillation who were deemed to be unsuitable for warfarin, the risk of hemorrhage was the reason in 23% of patients, but in another 50% the physician had considered warfarin to be ‘inappropriate’ [3].

Rationale for the use of prediction rules

Prediction of a high risk of bleeding has two main purposes. For patients in whom the risk of bleeding on anticoagulation is so high that it outweighs the risk of stroke, prophylaxis with an antiplatelet agent or perhaps no prophylaxis at all should be chosen. Initially, we used to compare the number of major bleeding events on warfarin with the number of ischemic stroke events without the treatment. Patients with a CHADS2 score (one point each for Congestive heart failure, Hypertension, Age over 75 years, and Diabetes, and two points for Stroke or transient ischemic attack) of ≥ 2 have an annual risk of stroke of at least 4% [4], which is higher than the annual risk of major bleeding on anticoagulation of 2–3% [5]. This subset was therefore the one that was recommended to receive anticoagulation. Lately, we have also started taking the burden of disease into account. As the majority of major bleeding events are extracranial [5], for which the burden is less than for an average stroke, even patients with atrial fibrillation and a single additional risk factor are now recommended to have anticoagulation for stroke prophylaxis [6, 7].

The second purpose of attempting to predict a high risk of bleeding is to minimize the risk situations while providing stroke prophylaxis for the patients at increased risk of hemorrhage. This could include elimination of concomitant therapies that may cause bleeding (antiplatelet agents, non-steroidal anti-inflammatory drugs, etc.), provision of intensified counseling against alcohol abuse, modification of risk of falls, treatment of gastrointestinal lesions, and last, but not least, extraordinarily careful supervision of anticoagulation. The last of these might include a combination of hemoglobin and prothrombin time monitoring, or referral to a specialized anticoagulation clinic.

Evaluation of a clinical prediction rule for bleeding

The prediction rules are usually established from a derivation cohort, and then validated in a different cohort. Ideally, the validation should be performed in cohorts with different indications, as patients with, for example, atrial fibrillation are, in several respects, different from those with venous thromboembolism (VTE) or those with mechanical heart valves. Patients assessed with a clinical prediction rule are typically categorized as having a high, intermediate or low risk of bleeding. The predictive value of these rules is often described with the c-statistic, which represents the area under the receiver operating characteristic curve, and where a value of 1.0 is absolute prediction and 0.5 is like tossing a coin. More interesting is the predictive value in those categorized as ‘high risk’, and this can be expressed as the likelihood ratio (LR). The LR should be at least 10 for a rule with strong performance; in a systematic review, none of four published prediction rules achieved this [8]. One prediction rule had an LR of 9, corresponding to a moderate performance [9].

The ultimate criterion for the performance is, however, when the prediction rules can reveal for which patients the risk of bleeding is so high that anticoagulation should be avoided, because the benefit of treatment is outbalanced by the harm of bleeding. With all of the current prediction rules, this situation only occurred for the combination of high risk of bleeding and atrial fibrillation with a CHADS2 score of 0 [10], but, for the latter condition, anticoagulation is not even recommended. Likewise, for patients with VTE, the rules did not identify a bleeding risk group for whom treatment should not be given.

Old age and hemorrhage

Old age, albeit with different cut-offs, is a component of every clinical prediction rule for risk of bleeding. According to a systematic review, the risk of bleeding doubles in the elderly [11]. The combination of old age and excessive prothrombin time seems to confer a particular increase in the risk of intracranial bleeding [12]. It should therefore be a priority to try and find a way to predict the risk in these elderly patients. In this issue of the journal, Scherz et al. [13] report on the evaluation of four of the published bleeding risk scores, using a cohort of patients aged ≥ 65 years who required anticoagulation because of acute VTE. The median age among the 663 patients was 75 years. The study was started in 2009, before the HAS-BLED score had been published [14]. The HEMORR2HAGES score could probably not be included because of the requirement for genetic data [15]. The c-statistic was between 0.49 and 0.60, and the positive LR for high vs. intermediate plus low risk was between 0.72 and 1.59. Thus, none of the prediction rules turned out to be helpful. Whether the subset of very elderly patients, i.e. ≥ 80 years, or the patients who are considered for extended anticoagulation (after the initial 3–6 months), would derive more benefit was not tested.

Implications for the management of VTE

The risk of bleeding in patients who are anticoagulated after VTE or for atrial fibrillation decreases quickly with time [5, 16]. During the initial period, when the risk of major bleeding might be as high as 3% per month [17], we do not have much of a choice, because the risk of recurrent VTE and even fatal pulmonary embolism without any anticoagulation is even higher [18]. It is after the strongly recommended 3–6 months of treatment that we need to weigh the risk of bleeding. At this time, the risk is, however, substantially lower, reaching 0.7–0.9% per year in contemporary trials [19-21]. For the patients reaching this point, we should probably use common sense rather than apply prediction rules.

Patients already showing a bleeding tendency should be considered for discontinuation of anticoagulation. Those with concomitant antiplatelet therapy should have the need for combined antithrombotic therapy reassessed, as warfarin also protects against myocardial infarction [22]. Patients with unstable prothrombin times should have their anticoagulant management and concomitant medications reviewed. Those with a history of falling should be worked up for amenable causes. Evidence of minor gastrointestinal or urogenital bleeding should prompt investigation to identify and possibly eliminate the source. Finally, the patients should be well informed about the symptoms of gastrointestinal bleeding and the need to seek medical attention without delay in the case of such an occurrence or for head trauma. The patients and family members also need to be aware of the fact that antithrombotic treatment is associated with an increased risk of major bleeding, but that the risk of serious thromboembolic events without treatment is even greater.

Disclosure of Conflict of Interests

  1. Top of page
  2. Disclosure of Conflict of Interests
  3. References

The author states that he has no conflict of interest.


  1. Top of page
  2. Disclosure of Conflict of Interests
  3. References
  • 1
    Budnitz DS, Lovegrove MC, Shehab N, Richards CL. Emergency hospitalizations for adverse drug events in older Americans. N Engl J Med 2011; 365: 200212.
  • 2
    Bungard TJ, Ghali WA, Teo KK, McAlister FA, Tsuyuki RT. Why do patients with atrial fibrillation not receive warfarin? Arch Intern Med 2000; 160: 416.
  • 3
    Connolly SJ, Pogue J, Hart RG, Hohnloser SH, Pfeffer M, Chrolavicius S, Yusuf S. Effect of clopidogrel added to aspirin in patients with atrial fibrillation. N Engl J Med 2009; 360: 206678.
  • 4
    Gage BF, Waterman AD, Shannon W, Boechler M, Rich MW, Radford MJ. Validation of clinical classification schemes for predicting stroke: results from the National Registry of Atrial Fibrillation. JAMA 2001; 285: 286470.
  • 5
    Schulman S, Beyth RJ, Kearon C, Levine MN. Hemorrhagic complications of anticoagulant and thrombolytic treatment: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th edition). Chest 2008; 133: 257S98S.
  • 6
    Camm AJ, Kirchhof P, Lip GY, Schotten U, Savelieva I, Ernst S, Van Gelder IC, Al-Attar N, Hindricks G, Prendergast B, Heidbuchel H, Alfieri O, Angelini A, Atar D, Colonna P, De Caterina R, De Sutter J, Goette A, Gorenek B, Heldal M, et al. Guidelines for the management of atrial fibrillation: the Task Force for the Management of Atrial Fibrillation of the European Society of Cardiology (ESC). Eur Heart J 2010; 31: 2369429.
  • 7
    You JJ, Singer DE, Howard PA, Lane DA, Eckman MH, Fang MC, Hylek EM, Schulman S, Go AS, Hughes M, Spencer FA, Manning WJ, Halperin JL, Lip GY. Antithrombotic therapy for atrial fibrillation: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-based Clinical Practice Guidelines. Chest 2012; 141: e531S75S.
  • 8
    Dahri K, Loewen P. The risk of bleeding with warfarin: a systematic review and performance analysis of clinical prediction rules. Thromb Haemost 2007; 98: 9807.
  • 9
    Beyth RJ, Quinn LM, Landefeld CS. Prospective evaluation of an index for predicting the risk of major bleeding in outpatients treated with warfarin. Am J Med 1998; 105: 919.
  • 10
    Holbrook A, Schulman S, Witt DM, Vandvik PO, Fish J, Kovacs MJ, Svensson PJ, Veenstra DL, Crowther M, Guyatt GH. 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: e152S84S.
  • 11
    Hutten BA, Lensing AW, Kraaijenhagen RA, Prins MH. Safety of treatment with oral anticoagulants in the elderly. A systematic review. Drugs Aging 1999; 14: 30312.
  • 12
    Fang MC, Chang Y, Hylek EM, Rosand J, Greenberg SM, Go AS, Singer DE. Advanced age, anticoagulation intensity, and risk for intracranial hemorrhage among patients taking warfarin for atrial fibrillation. Ann Intern Med 2004; 141: 74552.
  • 13
    Scherz N, Méan M, Limacher A, Righini M, Jaeger K, Beer H-J, Frauchiger B, Osterwalder J, Kucher N, Matter CM, Banyai M, Angelillo-Scherrer A, Lämmle B, Husmann M, Egloff M, Aschwanden M, Bounameaux H, Cornuz J, Rodondi N, Aujesky D. Prospective, multicenter validation of prediction scores for major bleeding in elderly patients with venous thromboembolism. J Thromb Haemost 2012; 11: 43543.
  • 14
    Pisters R, Lane DA, Nieuwlaat R, de Vos CB, Crijns HJ, Lip GY. A novel user-friendly score (HAS-BLED) to assess 1-year risk of major bleeding in patients with atrial fibrillation: the Euro Heart Survey. Chest 2010; 138: 1093100.
  • 15
    Gage BF, Yan Y, Milligan PE, Waterman AD, Culverhouse R, Rich MW, Radford MJ. Clinical classification schemes for predicting hemorrhage: results from the National Registry of Atrial Fibrillation (NRAF). Am Heart J 2006; 151: 71319.
  • 16
    Gomes T, Mamdani MM, Holbrook AM, Paterson JM, Hellings C, Juurlink DN. Rates of hemorrhage during warfarin therapy for atrial fibrillation. CMAJ 2013 Jan 21. Epub ahead of print.
  • 17
    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: 14452.
  • 18
    Barritt DW, Jordan SC. Anticoagulant drugs in the treatment of pulmonary embolism. Lancet 1960; 1: 130912.
  • 19
    Kearon C, Ginsberg JS, Kovacs MJ, Anderson DR, Wells P, Julian JA, MacKinnon B, Weitz JI, Crowther MA, Dolan S, Turpie AG, Geerts W, Solymoss S, van Nguyen P, Demers C, Kahn SR, Kassis J, Rodger M, Hambleton J, Gent M. Comparison of low-intensity warfarin therapy with conventional-intensity warfarin therapy for long-term prevention of recurrent venous thromboembolism. N Engl J Med 2003; 349: 6319.
  • 20
    Ridker PM, Goldhaber SZ, Danielson E, Rosenberg Y, Eby CS, Deitcher SR, Cushman M, Moll S, Kessler CM, Elliott CG, Paulson R, Wong T, Bauer KA, Schwartz BA, Miletich JP, Bounameaux H, Glynn RJ, Investigators ftP. Long-term, low-intensity warfarin therapy for the prevention of recurrent venous thromboembolism. N Engl J Med 2003; 348: 142534.
  • 21
    Schulman S, Eriksson H, Goldhaber S, Kakkar A, Kearon C, Kvamme AM, Mismetti P, Schellong S, Schnee J. Dabigatran or warfarin for extended maintenance therapy of venous thromboembolism. J Thromb Haemost 2011; 9 (Suppl. 2): 7312. (Abstr. O-Thu-033).
  • 22
    Hurlen M, Abdelnoor M, Smith P, Erikssen J, Arnesen H. Warfarin, aspirin, or both after myocardial infarction. N Engl J Med 2002; 347: 96974.