Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
Dr S E Vermeer
Correspondence to: Dr S E Vermeer, Erasmus Medical Center, Rotterdam, The Netherlands;
What are the frequency, risk factors, and consequences of magnetic resonance imaging (MRI)-defined silent brain infarcts?
Selected studies were original MRI studies of silent brain infarcts (ie, infarcts without overt stroke-like symptoms) in adults. For risk factor assessment, only population-based studies were included. Autopsy and computed tomography studies were excluded.
Outcomes were frequency, risk factors, and consequences of silent brain infarcts.
PubMed (1966 to November 2006) and reference lists of relevant studies were searched for English-language studies. 105 articles met the selection criteria, including 24 reports on risk factors (21 cross-sectional and 3 longitudinal, n = 128–3397, mean age range 59–75 y) from 8 population-based cohorts.
Prevalence of silent brain infarcts in 8 population-based cohorts was between 8% and 28%, increasing with age (table). Age (11 reports) and hypertension (5 reports) were most often associated with silent infarcts, although 3 reports did not report hypertension as a risk factor. Other risk factors identified in those cohorts included severe white matter lesions (4 reports); and cardiovascular disease, creatinine concentrations, or retinal microvascular abnormalities (3 reports each). Risk of infarct was not increased based on sex (8 reports); cholesterol concentrations (6 reports); diabetes or smoking (5 reports each); alcohol use (4 reports); or fibrinogen or C-reactive protein concentrations (3 reports each). In 2 population-based cohorts (n = 4474), silent brain infarcts were associated with a 2–4 fold increase in risk of subsequent stroke, independent of cardiovascular risk factors, and in 1 cohort (n = 1077) infarcts were associated with double the risk of dementia.
Silent brain infarcts occur in 8–28% of the general population and are associated with subsequent stroke and dementia.
Vermeer SE, Longstreth WT Jr, Koudstaal PJ. Silent brain infarcts: a systematic review. Lancet Neurol 2007;6:611–9.
▸Clinical impact ratings: IM/Ambulatory care 6/7; Neurology 6/7
Small, deep infarcts on MRI are often silent and, as shown in the review by Vermeer et al, surprisingly frequent. Differentiating small, deep lesions histologically is often difficult, so it is not surprising that lacunar infarcts, focal ischaemic rarefaction, gliosis, and perivascular demyelination may have similar MRI appearances.1
The criteria for silent infarcts in the studies reviewed by Vermeer et al varied widely. Focal lesions ⩾3 mm were considered lacunar infarcts in most studies; however, such a cutpoint would allow many larger enlarged perivascular spaces (EPVSs) to be counted as infarcts.2 Proton density/fluid-attenuated inversion recovery (FLAIR) sequences are necessary to differentiate EPVSs from infarcts. The high rate of silent infarcts seen in some of the studies in the review may relate to the inclusion of non-infarct and even non-ischaemic lesions.
Despite this, infarcts identified by MRI did show an expected increase with age and at least a doubling of incident stroke in 2 population-based cohorts.3 4 Although most silent infarcts are lacunar,4 that classification applied to only 6% of incident strokes,3 which is surprising. In patients with silent infarcts, atrial fibrillation was a risk factor for incident stroke in 1 study,3 particularly if the MRI infarct was cortical, which suggests that identification of cardioembolic sources is important.
Standardisation of MRI criteria for identifying silent lesions and further population studies are needed to confirm the increased risk, subtype, and severity of incident stroke and risk of dementia associated with silent infarcts. The evidence is not clear on what to do clinically about silent infarcts; however, when single cortical or multiple subcortical infarcts are found, stroke risk factors (eg, atrial fibrillation) should be evaluated and managed appropriately.
Source of funding: no external funding.