International Stroke Conference
January 24–26

Abbas Kharal, MD, MPH

The topic of whether or not anticoagulation should be restarted after an anticoagulation-related lobar intracerebral hemorrhage (ICH) remains a hot debate among neurologists around the world. Although with the advent of better imaging-based risk predictors for lobar ICH, including cerebral microbleeeds and sulcal siderosis on MRI, raising concern for cerebral amyloid angiopathy and posing higher recurrent ICH risks, there are also accordingly more safer options available now for oral anticoagulation, e.g. direct thrombin inhibitors and surgical alternatives like left atrial appendage closure, which may help lower mitigate the risks of recurrent lobar intracerebral hemorrhage. However, there still remains insufficient data to help definitively guide our management decisions when deciding whether or not it is truly safe to resume anticoagulation in such high-risk patients with a prior anticoagulation-related intracerebral hemorrhage.

An interesting crossfire debate was held on this very topic at the International Stroke Conference’s closing event between Dr. Stephan Mayer from Henry Ford Hospital, who spoke glamorously in favor of not resuming AC in such patients, and Dr. Alessandro Biffi from Massachusetts General Hospital, a world renowned cerebrovascular epidemiologist who has published extensively on the topic of intracerebral hemorrhage and anticoagulation risks, who spoke against the notion of holding oral anticoagulation in all patients with AC-related lobar ICH. Dr. Mayer raised concerns about the high risk of recurrent ICH being approximately 10.4% per year based on previously published data from Dr. Biffi and colleagues, which, when paired with the presumed increase in mortality associated with it from previously published data, appears to outweigh any potential benefits of resuming anticoagulation in such patients. Dr. Mayer went on to conclude that resuming anticoagulation in such patients would be “nuts!”

However, persuasively arguing against the notion of holding anticoagulation in such patients with AC-related lobar ICH indefinitely, Dr. Biffi eluded to pooled data analysis from three prior studies, including studies by Nielson et al (Circulation 2015)1, Murthy et al (Stroke 2017)2, and Biffi et al. (Ann. Neurol 2017)3, in which 621/1752 (35%), 786/2044 (38%) and 87/379 (23%) patients, respectively, had an anticoagulation-related lobar ICH and were restarted on oral anticoagulation (OAC) due to the presence of pre-existing high-risk thromboembolic conditions necessitating resumption of AC. He further delved into the data recently published from his group in which the 23% patients whom had their OAC resumed were noted to have a statistically significant reduction of 49% in All-Strokes (HR 0.51 CI 0.32 – 0.8, p = 0.004), i.e. both ischemic stroke and recurrent ICH, and although there was a slight non-significant increase in recurrent ICH (HR 1.12, CI 0.94 – 1.34, p=0.22), the overall rate of mortality was significantly lower (HR 0.30, p=0.037) in the AC resumption group. Most notably, patients with probable CAA in the AC resumption group also had a significantly higher rate of favorable outcomes (defined as mRS 0-3) at 1 year when compared to those in whom AC was held indefinitely (HR 3.11, p= 0.038).

Although the issue of whether or not it is safe to resume oral anticoagulation in patients with AC-related lobar ICH remains a controversial topic, the pooled data analysis presented by Dr. Biffi emphasizes that perhaps there may be a subgroup of patients with higher thromboembolic risks that may significantly benefit in the long run from resuming oral anticoagulation. Although holding anticoagulation indefinitely for some high-recurrent ICH risk patients would make plausible sense, however, resuming AC in patients with lower ICH risk, including those with low cerebral microbleed burden, no (or perhaps only focal) cortical superficial siderosis and a high thromboembolic risk, e.g. AF with higher CHADS2VASc score, may be reasonable. Such subgroup of patients may benefit in the long run in terms of lower mortality and stroke recurrence and a significantly better functional outcome, based on the recently presented data. Remarkably, there was a 75% agreement from audience polls in favor of resuming OAC in such subgroup of patients.

Nonetheless, it is important to acknowledge that such pooled data analysis may be highly prone to an unmeasurable selection bias and possible confounding by indication, a point that Dr. Mayer also mentioned in his rebuttal. We also do not know exactly what oral anticoagulant agents each of those patients were started on, what the timing of resumption was and whether or not there may be a confounding by indication in the studies used for pooled analysis as many physicians may have avoided vitamin K antagonists (VKA) given the higher risk of ICH associated with VKA. Whether or not it is safe to resume OAC after an AC-related ICH, if safe, then which subgroup of patients is it safest to resume OAC in, within what time frame would it be safest to resume their AC, and which AC agents would be safest in such high-risk patients still remains an area that requires further research for which a randomized controlled clinical trial is emphatically needed.

References:

  1. Nielsen PB, Larsen TB, Skjøth F, Gorst-Rasmussen A, Rasmussen LH, Lip GYH. Restarting Anticoagulant Treatment After Intracranial Hemorrhage in Patients With Atrial Fibrillation and the Impact on Recurrent Stroke, Mortality, and Bleeding: A Nationwide Cohort Study. Circulation [Internet] 2015 [cited 2018 Jan 28];132(6):517–25. Available from: http://www.ncbi.nlm.nih.gov/pubmed/26059010
  2. Murthy SB, Gupta A, Merkler AE, et al. Restarting Anticoagulant Therapy After Intracranial Hemorrhage. Stroke [Internet] 2017 [cited 2018 Jan 28];48(6):1594–600. Available from: http://www.ncbi.nlm.nih.gov/pubmed/28416626
  3. Biffi A, Kuramatsu JB, Leasure A, et al. Oral Anticoagulation and Functional Outcome after Intracerebral Hemorrhage. Ann Neurol [Internet] 2017 [cited 2018 Jan 28];82(5):755–65. Available from: http://doi.wiley.com/10.1002/ana.25079