Background and Purpose-We assessed whether the presence, number, and distribution of cerebral microbleeds (CMBs) on pre-intravenous thrombolysis MRI scans of acute ischemic stroke patients are associated with an increased risk of intracerebral hemorrhage (ICH) or poor functional outcome. Methods-We performed an individual patient data meta-analysis, including prospective and retrospective studies of acute ischemic stroke treated with intravenous tissue-type plasminogen activator. Using multilevel mixed-effects logistic regression, we investigated associations of pre-treatment CMB presence, burden (1, 2-4, ≥5, and >10), and presumed pathogenesis (cerebral amyloid angiopathy defined as strictly lobar CMBs and noncerebral amyloid angiopathy) with symptomatic ICH, parenchymal hematoma (within [parenchymal hemorrhage, PH] and remote from the ischemic area [remote parenchymal hemorrhage, PHr]), and poor 3-to 6-month functional outcome (modified Rankin score >2). We performed a large-scale pooled individual patient data meta-analysis of quality observational studies to test the following hypotheses: (1) there is a relationship between increasing CMB burden and ICH risk 8,9 ; (2) strictly lobar CMBs (reflecting possible or probable cerebral amyloid angiopathy [CAA]) and mixed or strictly deep CMBs (likely associated with hypertensive arteriopathy) have different effects on ICH risk; (3) CMBs are associated more strongly with the risk of remote ICH than other ICH types 10 ; and (4) CMBs are associated with worse functional outcome.
Results-In
Methods
Study Design and Inclusion CriteriaWe identified prospective or retrospective studies that assessed pretreatment MRI-defined CMBs, ICH, and 3-to 6-month functional outcome after acute ischemic stroke, treated solely with intravenous tPA from a systematic review prepared according to Preferred Reporting Items for Systematic Review and Meta-Analyses 7,11 (updated August 1, 2015). We searched PubMed for micro(-)bleed*, or micro(-)h(a) emorrhag*, or gradient-echo, or susceptibility-weighted in association with thromboly* or tPA, or tissue plasminogen activator 7 ; reference lists; and authors' own files. Figure I in the online-only Data Supplement shows a flow diagram.We collected anonymized individual patient detailed clinical data and CMB counts in lobar, deep, and infratentorial regions according to standardized definitions 6,12,13 using standardized report forms. A prespecified protocol was circulated to collaborators but not published.
OutcomesWe defined ICH according to ECASS-2 (European Cooperative Acute Stroke Study II), 14,15 including hemorrhagic infarction, parenchymal hemorrhage (PH), and sICH 16 (acute intracerebral blood and associated increase in National Institutes of Health Stroke Scale ≥4 points, except 1 study 17 that used the definition in the PROACT-II trial [Prolyse in Acute Cerebral Thromboembolism]).18 Remote parenchymal hemorrhage (PHr) was defined as ICH remote from the symptomatic ischemic area. 10 We defined poor outcome at 3 to 6 months as modified...