Antiplatelets after intracerebral haemorrhage: treat the patient, not the brain imaging

Clinicians underestimate harms and overestimate benefits from medical interventions.1Hoffmann TC Del Mar C Clinicians' expectations of the benefits and harms of treatments, screening, and tests: a systematic review.JAMA Intern Med. 2017; 177: 407-419Crossref PubMed Scopus (260) Google Scholar However, according to a report by Rustam Al-Shahi Salman and colleagues2Al-Shahi Salman R Minks DP Mitra D et al.Effects of antiplatelet therapy on stroke risk by brain imaging features of intracerebral haemorrhage and cerebral small vessel diseases: subgroup analyses of the RESTART randomised, open-label trial.http://dx.doi.org/10.1016/S1474-4422(19)30184-XGoogle Scholar published in The Lancet Neurology, the perceived risk from restarting antiplatelet therapy after intracerebral haemorrhage in patients with cerebral microbleeds has been substantially overestimated on the basis of observational data. Use of antiplatelet therapy in patients with intracerebral haemorrhage is common,3Thompson BB Béjot Y Caso V et al.Prior antiplatelet therapy and outcome following intracerebral hemorrhage: a systematic review.Neurology. 2010; 75: 1333-1342Crossref PubMed Scopus (141) Google Scholar a conservative estimate is one in four patients, so more certainty in making decisions regarding restarting such therapy is highly relevant. The RESTART trial,4RESTART CollaborationEffects of antiplatelet therapy after stroke due to intracerebral haemorrhage (RESTART): a randomised, open-label trial.Lancet. 2019; (published online May 22.)http://dx.doi.org/10.1016/S0140-6736(19)30840-2Google Scholar which is published in The Lancet, randomly assigned 537 survivors of intracerebral haemorrhage that occurred while taking antithrombotic therapy to start or avoid antiplatelet therapy. The investigators reported no significant differences in risk of recurrent intracerebral haemorrhage but a reduced risk, albeit not reaching statistical significance, of both haemorrhagic and ischaemic stroke was observed consistently in subgroups in patients allocated to start antiplatelet therapy.4RESTART CollaborationEffects of antiplatelet therapy after stroke due to intracerebral haemorrhage (RESTART): a randomised, open-label trial.Lancet. 2019; (published online May 22.)http://dx.doi.org/10.1016/S0140-6736(19)30840-2Google Scholar The neuroimaging substudy2Al-Shahi Salman R Minks DP Mitra D et al.Effects of antiplatelet therapy on stroke risk by brain imaging features of intracerebral haemorrhage and cerebral small vessel diseases: subgroup analyses of the RESTART randomised, open-label trial.http://dx.doi.org/10.1016/S1474-4422(19)30184-XGoogle Scholar of the RESTART trial, published in The Lancet Neurology, focused on the 254 patients with MRI performed before randomisation. By contrast with previous observational data, there were no indications that starting antiplatelet therapy increased risk of recurrent intracerebral haemorrhage in patients with cerebral microbleeds or superficial siderosis. The number of microbleeds did not seem to affect risk of recurrent intracerebral haemorrhage. RESTART is the first randomised trial to address this subject; until now, data were based on observational hospital cohorts that suggested excess risk of recurrent intracerebral haemorrhage from reintroduction of antiplatelets, with the highest risk in patients with cerebral microbleeds.5Biffi A Halpin A Towfighi A et al.Aspirin and recurrent intracerebral hemorrhage in cerebral amyloid angiopathy.Neurology. 2010; 75: 693-698Crossref PubMed Scopus (246) Google Scholar Uncertainty remains in relation to the precision of these findings. RESTART was underpowered: a sample size of 2200 is suggested by the investigators as needed for a potentially conclusive study. However, findings were consistent across prespecified subgroups. Selection bias towards small haematomas was present: median intracerebral haemorrhage volume was 3·7 mL in the intervention group and 4·3 mL in the control group, and few participants had high numbers of microbleeds. There was a predominance of men in the MRI substudy and entire cohort. Also, all MRI field strengths were accepted and gradient-recalled echo was used for bleeding detection, which most likely led to lower detection rates of cerebral microbleeds than what would have been the case if based on high field strengths or susceptibility weighted imaging, or both.6Nandigam RN Viswanathan A Delgado P et al.MR imaging detection of cerebral microbleeds: effect of susceptibility-weighted imaging, section thickness, and field strength.AJNR Am J Neuroradiol. 2009; 30: 338-343Crossref PubMed Scopus (316) Google Scholar Consequently, the results could be substantially different in an adequately powered trial and therefore the external validity of the results of the current substudy in relation to the overall intracerebral haemorrhage population is unclear—for example, these results cannot be directly extrapolated to patients with larger haematoma volumes or large numbers of microbleeds. However, the investigators have succeeded in recruiting a high-risk population; the proportion of recurrent intracerebral haemorrhage was 4·9% in the start antiplatelet therapy group versus 9·1% in the control group during an average follow-up of 2 years.2Al-Shahi Salman R Minks DP Mitra D et al.Effects of antiplatelet therapy on stroke risk by brain imaging features of intracerebral haemorrhage and cerebral small vessel diseases: subgroup analyses of the RESTART randomised, open-label trial.http://dx.doi.org/10.1016/S1474-4422(19)30184-XGoogle Scholar Furthermore, consideration of re-initiation of antiplatelet therapy is most pertinent in patients with small haematomas and good outcome. These data are indeed surprising because most stroke clinicians would have expected that presence of cerebral microbleeds heralds higher risk of recurrent intracerebral haemorrhage on starting antiplatelet therapy, because the higher risk of intracerebral haemorrhage in patients with microbleeds has been well documented,7Charidimou A Boulouis G Roongpiboonsopit D et al.Cortical superficial siderosis and recurrent intracerebral haemorrhage risk in cerebral amyloid angiopathy: large prospective cohort and preliminary meta-analysis.Int J Stroke. 2019; (published online Feb 20.)DOI:10.1177/1747493019830065PubMed Google Scholar and observational data support further increase in risk with use of antithrombotic drugs.5Biffi A Halpin A Towfighi A et al.Aspirin and recurrent intracerebral hemorrhage in cerebral amyloid angiopathy.Neurology. 2010; 75: 693-698Crossref PubMed Scopus (246) Google Scholar, 8Charidimou A Karayiannis C Song TJ et al.Brain microbleeds, anticoagulation, and hemorrhage risk: meta-analysis in stroke patients with AF.Neurology. 2017; 89: 2317-2326Crossref PubMed Scopus (70) Google Scholar That antiplatelet therapy increases risk of intracerebral haemorrhage overall and that patients after intracerebral haemorrhage have high risk of recurrence do not necessarily imply that antiplatelet therapy will increase risk of intracerebral haemorrhage in this patient population. Furthermore, imaging findings such as the presence of microbleeds can often arouse concern, which can be worrying even in the absence of intracerebral haemorrhage—but possibly more for the clinician than for the patient's prognosis. These findings underline that our pathophysiological understanding remains incomplete and we should not rely on observational data, but always aim for randomised controlled data. More data are certainly needed in this field, including data on specific antiplatelet drugs and specific combinations of antiplatelet drugs that are frequently used—eg, standard regimens for acute coronary syndrome. Data from randomised controlled trials are also needed for oral anticoagulants, for which a parallel discussion is ongoing. My clinical interpretation of these results is that physicians can be less restrictive in prescribing antiplatelets to patients with minor intracerebral haemorrhage. Decisions cannot be made on the basis of MRI findings such as cerebral microbleeds; according to these data, there is no excess risk attributed to the antiplatelet intervention in this population, so the focus of a therapeutic decision should rest not on the reasons to prescribe antiplatelets, but on the reasons not to prescribe them. However, the best evidence remains to be from randomised trials when available: RESTART-Fr (NCT02966119) and STATICH (NCT03186729) are ongoing and might settle any remaining uncertainty. I declare no competing interests. Effects of antiplatelet therapy on stroke risk by brain imaging features of intracerebral haemorrhage and cerebral small vessel diseases: subgroup analyses of the RESTART randomised, open-label trialOur findings exclude all but a very modest harmful effect of antiplatelet therapy on recurrent intracerebral haemorrhage in the presence of cerebral microbleeds. Further randomised trials are needed to replicate these findings and investigate them with greater precision.