Cognitive impairment before intracerebral hemorrhage is associated with cerebral amyloid angiopathy

Banerjee, Gargi and Wilson, Duncan and Ambler, Gareth and Appiah, Karen Osei Bonsu and Shakeshaft, Clare and Lunawat, Surabhika and Cohen, Hannah and Yousry, Tarek and Habil, Med and Lip, Gregory Y.H. and Muir, Keith W. and Brown, Martin M. and Salman, Rustam Al Shahi and Jäger, Hans Rolf and Werring, David J. and Shaw, Louise and Sword, Jane and Nor, Azlisham Mohd and Sharma, Pankaj and Veltkamp, Roland and Kelly, Deborah and Harrington, Frances and Randall, Marc and Smith, Matthew and Mahawish, Karim and Elmarim, Abduelbaset and Esisi, Bernard and Cullen, Claire and Nallasivam, Arumug and Price, Christopher and Barry, Adrian and Roffe, Christine and Coyle, John and Hassan, Ahamad and Lovelock, Caroline and Birns, Jonathan and Cohen, David and Sekaran, L. and Parry-Jones, Adrian and Parry, Anthea and Hargroves, David and Proschel, Harald and Datta, Prabel and Darawil, Khaled and Manoj, Aravindakshan and Burn, Mathew and Patterson, Chris and Giallombardo, Elio and Smyth, Nigel and Emsley, Hedley (2018) Cognitive impairment before intracerebral hemorrhage is associated with cerebral amyloid angiopathy. Stroke, 49 (1). pp. 40-45. ISSN 0039-2499

Abstract

Background and Purpose-Although the association between cerebral amyloid angiopathy (CAA) and cognitive impairment is increasingly recognized, it is not clear whether this is because of the impact of recurrent intracerebral hemorrhage (ICH) events, disruptions caused by cerebral small vessel damage, or both. We investigated this by considering whether cognitive impairment before ICH was associated with neuroimaging features of CAA on magnetic resonance imaging. Methods-We studied 166 patients with neuroimaging-confrmed ICH recruited to a prospective multicentre observational study. Preexisting cognitive impairment was determined using the Informant Questionnaire on Cognitive Decline in the Elderly (IQCODE). Magnetic resonance imaging markers of cerebral small vessel disease, including CAA, were rated by trained observers according to consensus guidelines. Results-The prevalence of cognitive impairment before ICH was 24.7% (n=41) and, in adjusted analyses, was associated with fulflling the modifed Boston criteria for probable CAA at presentation (odds ratio, 4.01; 95% confdence interval, 1.53-10.51; P=0.005) and a higher composite CAA score (for each point increase, odds ratio, 1.42; 95% confdence interval, 1.03-1.97; P=0.033). We also found independent associations between pre-ICH cognitive decline and the presence of cortical superfcial siderosis, strictly lobar microbleeds, and lobar ICH location, but not with other neuroimaging markers, or a composite small vessel disease score. Conclusions-CAA (defned using magnetic resonance imaging markers) is associated with cognitive decline before symptomatic ICH. This provides evidence that small vessel disruption in CAA makes an independent contribution to cognitive impairment, in addition to effects due to brain injury caused directly by ICH.