Qing Hao, MD
With the recent advance in image techniques, MRI has helped tremendously in understanding the mechanism of cerebrovascular disease. “The more you know, the more you know you don’t know” (Aristotle). When patients not only have a TIA or ischemic stroke, but also cerebral microbleeds (CMB), what should we do with the antiplatelet agent?
Aiming to find some solutions for this treatment dilemma, Lau, Lovelock and colleagues prospectively studied 2156 patients with diagnosis of TIA/ischemic stroke from two large cohorts in the United Kingdom and Hong Kong, respectively, to study the risks and time course of recurrent ischemic events, ICH/extracranial hemorrhage stratified by microbleeds burden.
On the west side, Oxford Vascular Study(OXVASC) is an ongoing population-based study of all acute vascular events occurring within a predominantly white population of all 92,728 individuals who are registered with 100 general practitioners in 8 general practices of Oxfordshire, United Kingdom; 1080 consecutive TIA/ischemic stroke patients with MRI available (CMB was assessed using GER on 1.5-T or 3-T scanner) from 2004 to 2014 were included in this analysis. On the east side, 1076 consecutive patients who were predominantly Chinese with a diagnosis of ischemic stroke and MRI available (CMB was assessed using SWI on 3-T scanner) from 2008 to 2014 were included in the analysis. CMB were defined by 2 consultant neuroradiologists using Microbleed Anatomical Rating Scale and burden graded as absent, 1, 2 to 4, and ≥5.
Among 1811 patients who were prescribed with antiplatelet therapy, the 5-year risks of recurrent ischemic stroke and ICH both increased as the number of CMB increases. In multivariable analysis, high microbleed burden (≥5) was an independent predictor of recurrent ischemic stroke, ICH, all cause mortality, and nonvascular death. The pooled results from these two cohorts with those from a recent meta-analysis showed that among patients taking antiplatelet agents, compared with patients with no CMB, those with ≥ 5 CMB were at 3- fold higher risk of recurrent ischemic stroke and 13- fold increased risks of ICH.
For the patients with <5 CMB, the 5-year absolute risk of ischemic stroke was much higher than ICH, including both non-disabling and disabling events. For those with ≥5 CMB, the risk of disabling/fetal ICH was similar to the disabling/fetal ischemic event.
In the time-course analysis, for the patients with <5 CMB, the risk of combined ischemic events (stroke and coronary events) is higher than the risk of combined hemorrhagic events (ICH and major extracranial bleed) both within 1 year and during year 2-5. For those with 5≥CMB, risk of combine ischemic events is higher than the combined hemorrhagic events within the first year, but they were similar during year 2-5.
These findings provide important insights when facing the dilemmas as aforementioned:
- Should we give antiplatelet to TIA or stroke patients if they have microbleeds? Yes, because microbleed is associated with higher risks of ischemic events than hemorrhagic events during the first year regardless of the number of the microbleeds. This study, however, is unable to examine the outcome stratified by the location of CMB. Of note, patients with transient neurological deficits and diagnosis of cerebral amyloid angiopathy were excluded; patients with higher mRS at discharge (possibly more severe stroke) were excluded as well.
- Should we individualize the antiplatelet therapy duration based on the number of microbleeds? Yes.
a. For patients with <5 microbleeds, risk of ischemic strokes is always higher than hemorrhagic stroke, so antiplatelet may be continued lifelong. But you may ask, How about if the patient has 1 or 2 more new MCB during their follow up images?
b. For patients with ≥5 microbleeds, due to increasing longer-term risks of ICH that are more likely to be the disabling/fetal ones, gradual discontinuation of antiplatelet after the first year may be considered. However, to draw a definitive conclusion, we need clinical trials to compare the longer-term outcome in patients with ≥5 microbleeds who are on short- vs. long-term antiplatelet therapy.
Thank you for this great summary. Is there going to be additional follow up reports on the effect of CMB location (lobar vs deep)? And do you think the threshold (here more than 5) can be different based on magnet strength (1.5T vs more recently 3T)? It will also be interesting and important to know the dosing, and effect of non-ASA anti platelets, as well as when to restart treatment after a hemorrhagic event.