35 The epidemiology and symptomology of functional stroke mimics: a systematic review and meta-analysis

Abstract

Aims Reconfiguration of stroke services in England has emphasised fast diagnosis and treatment and subsequently, the proportion of stroke mimic patients entering stroke pathways has been highlighted. Stroke mimic patients may be ‘medical mimics’, with medical explanations for symptoms, e.g. syncope, seizure, but a proportion of presenting patients have a functional aetiology. Functional stroke mimics accounted for 8% of admissions to an acute stroke service in London (Gargalas et al., 2015) and the prevalence and burden of functional stroke is well recognised by clinicians. We present a systematic review and meta-analysis aiming to: 1) estimate the prevalence of stroke mimics and functional stroke mimics across medical settings; and 2) describe the demographic and symptom profiles of functional stroke patients. Methods Three literature searches took place between 2015–2018 utilising OvidSP, PubMed, CINAHL and Google Scholar. A total of 13 974 abstracts were reviewed and 114 papers met inclusion criteria. Age and sex proportions were compared between stroke, stroke mimic and functional mimic groups. Prevalence rates across settings and moderators of functional mimic rates were calculated using random-effects models. Results Stroke, stroke mimic and functional mimic definitions varied between studies. Across settings, 25% of suspected stroke patients were stroke mimics and 15% of stroke mimics had a functional aetiology. Stroke mimics were younger than stroke patients and more likely to be female. Similarly, functional patients were younger and more often female than medical mimics. 10 papers gave symptom information for functional patients; compared to medical mimics, functional patients were more likely to display weakness/numbness and less likely to present with reduced consciousness, visual symptoms or speech/language symptoms. Meta-analyses show a higher rate of stroke mimics in primary care (38%) vs more acute settings (12%) but the inverse for functional mimics (24% in stroke units vs only 12% in primary care). Functional rates were highest in studies that were descriptive, retrospective, from high income countries and in studies where all patients received thrombolysis. Conclusions Functional diagnoses are an important differential of suspected stroke. Definitions of functional stroke mimics vary widely in stroke literature. Our findings suggest functional stroke patients are most commonly seen in tertiary settings. There are no guidelines on the management of these patients within acute stroke settings. In the context of these findings, a feasibility study is underway investigating the presentation of functional stroke patients and their views on possible interventions and this research may help improve current care pathways.