Relationship between silent cerebral infarcts and quality of anticoagulation in patients with prosthetic mitral valves.

Abstract

OBJECTIVE\nAlthough patients with prosthetic heart valves have an increased risk of clinically overt cerebrovascular events, evidence for the risk of silent cerebral infarction (SCI) is scarce. Serum neuron-specific enolase (NSE) is suggested to be a valid biomarker that allows for the quantification of the degree of neuronal injury. We aimed to assess whether NSE is elevated as a marker of recent SCI in patients with a prosthetic mitral valve.\n\n\nMETHODS\nWe measured the NSE levels in 103 patients with a prosthetic mitral valve (PMV), admitted to our outpatient clinics for routine evaluation. International normalized ratio (INR) and time in target therapeutic range (TTR) were noted as anticoagulation quality measures.\n\n\nRESULTS\nMost of the patients were females (58%), and a mean age was 65 years. NSE values of >12 ng/mL, suggesting a recent SCI, was detected in 25 patients (24%). NSE was negatively correlated with admission INR (r=-0.307, p=0.002). Multivariate analyses demonstrated subtherapeutic INR (INR <2.5) and suboptimal TTR as independent predictors of SCI [odds ratio (OR) 5.420; 95% confidence interval (CI) 1.589 to 18.483; p=0.007, and OR 4.149; 95% CI 1.019 to 16.949; p=0.047, respectively]. Being a current smoker (OR 10.798; 95% CI 2.520 to 46.272; p=0.001), large left atrium (OR 6.763; 95% CI 2.253 to 20.302; p=0.001), and not using aspirin (OR 10.526; 95% CI 1.298 to 83.333; p=0.027) were other independent predictors.\n\n\nCONCLUSION\nOur data suggest that silent brain infarcts are very prevalent among patients with a PMV, as one fourth of them had the event during their routine outpatient visit. Poor quality of anticoagulation partly explains the increased prevalence.