P175 Burden of cerebral infarcts identified by screening cerebral mri scans in patients with pulmonary arteriovenous malformations

Introduction In a recent UK study, 61/497 (12.3%) of consecutive patients with pulmonary arteriovenous malformations (PAVMs) had experienced a clinical ischaemic stroke at median 46 (range 16–82) years. 1 Conventional stroke management included antiplatelet agents, but since many PAVM patients have underlying hereditary haemorrhagic telangiectasia (HHT), there has not been a blanket recommendation to treat all PAVM patients with such agents if residual PAVMs remain after maximal treatment. The goal of this study was to evaluate evidence of silent ischemia in patients with PAVMs. Methods Between 20/04/2009 and 02/12/2016, 43 individuals (20 males; 23 females) with known or suspected HHT underwent a cerebral MRI scan performed for the purpose of HHT cerebral AVM screening. All available scans were analysed by two independent neuroradiologists, blinded to patient demographics/PAVM status. Data were subsequently categorised and analysed using STATA IC v13 (Statacorp, Texas). Results Patient ages ranged from 17–74 (mean 42.2) years. Twenty-two (51.1%) were known to have PAVMs demonstrable by thoracic CT scan, and 21 had PAVMs excluded by CT scan. There was no age difference between the PAVM and non-PAVM cohorts (mean 43.3 [range 16–73] versus 41 [21–65] years respectively, p=0.46). No scan demonstrated a cerebral AVM, none provided evidence of prior cerebral haemorrhage, but only 22 (51.2%) of scans were reported as normal. 17 (81%) of patients without PAVMs had a normal scan, compared to 5 (22.7%) PAVM patients (p=0.0002). 15/21 (68.2%) PAVM patients had at least one infarct, and 6/21 (27.3%) had microangiopathic changes. The mean number of infarcts per PAVM patient was 1.58 compared to 0.14 in non-PAVM patients (p Conclusions The findings identify high rates of silent cerebral ischaemic changes in patients with PAVMs, and raise the question whether all patients with persistent PAVMs after treatment should have pharmacological stroke prevention therapy, in the absence of a clinical stroke. Reference Shovlinet al. PLOS One2014, Feb 19;9(2):e88812.