IMAGING BIOMARKERS OF CEREBRAL SMALL VESSEL DISEASE IN ADULTS WITH CONGENITAL HEART DISEASE

Aim Imaging biomarkers in medical fields other than oncology tend to be confined to research settings. In neuroimaging, biomarkers of cerebral small vessel disease (SVD) have shown promising results for translation into the clinic but need further validation. In this work, we strove for a more clinically-oriented classification of one of the most representative of SVD biomarkers, i.e. white matter hyperintensities (WMHs), on a large cohort of community-dwelling subjects. We also used this classification to better analyse signs of SVD in a selected population of adults with congenital heart disease. Lastly, we evaluated the steps to be taken for achieving standardisation of WMH assessment. Cerebral small vessel disease in adults with tetralogy of Fallot: preliminary results In this section we describe our exploratory research on a small sample of 10 adult patients with surgically corrected tetralogy of Fallot and 10 age- and sex-matched control subjects. Cerebral microbleeds were visible in every patient and in one control subject. Also, a significant correlation between WMH volume and severity of symptoms of heart failure was observed. Even though the aetiology of the observed findings remained obscure, cerebral microbleeds and WMHs may indicate an increased susceptibility to brain microvascular damage in congenital heart disease and have been deemed worthy of further investigations. Imaging biomarkers of small vessel disease in adults with congenital heart disease: a systematic review Considering our preliminary findings, we extensively reviewed the available literature on vascular injury in adults with congenital heart disease. We found few preliminary studies on this topic, all of which carried several methodological limitations. However, all studies reported a variety of brain vascular changes in the examined patients, ranging from signs of SVD to silent stroke and cortical atrophy. The assumption that these findings differ from those visible in children with heart defects needs to be confirmed. For this purpose, neuroimaging studies in adults with congenital heart disease are needed to differentiate past global lesion burden from present chronic ongoing cerebrovascular disease. Automatic sub-classification of white matter hyperintensities: application to a large community-dwelling cohort As WMH total volume is variably associated with cognition, we developed an automatic method to classify them into four categories according to lesion location (periventricular versus deep) and lesion intensity in T1 and T2-weighted sequences. We applied this method on brain scans from 684 older adults from the Whitehall II study. We also showed that periventricular white matter hyperintensities that appear hypointense in T1-weighted images were significantly associated with poorer performance in several cognitive tests in this cohort. Interestingly, we found no association between total WMH volume and cognition. These findings suggest that sub-classifying WMHs according to both location and intensity in T1-weighted images provides added value when studying the clinical correlates of this imaging biomarker. Inconsistency in quantifying and reporting white matter hyperintensities volume: a systematic review We tried to estimate a normative range for WMH volume in healthy ageing using the highest level of evidence. We meta-analysed 2743 healthy subjects’ WMH volume reported from 17 studies. Our results showed an extremely high heterogeneity across the examined studies. We thereby proposed methodological strategies needed to overcome the current inconsistency in WMH assessment, such as harmonisation of image acquisition and standardisation of anatomical definitions. Most importantly, effective communication between researchers and clinicians is strongly warranted to translate technical know-how for imaging biomarkers assessment into clinical practice. The BACH Study In view of our preliminary results and considering the dearth of scientific evidence on the topic of brain involvement in adults with congenital heart disease, we launched the Brain Aging in Congenital Heart disease (BACH) San Donato study. This multidisciplinary study comprises an extensive brain imaging protocol to investigate signs of small vessel disease and a thorough neuropsychological battery to test patients’ cognitive performance. We found that automatically-detected white matter hyperintensities located in the deep white matter were associated with poorer performance at the frontal assessment battery. Also, this study confirmed that patients had a much larger number of cerebral microbleeds than healthy controls. Conclusions In this thesis we showed that cerebral microbleeds are over-expressed in adult patients with congenital heart disease. Longitudinal studies will aid in clarifying the role of this and other biomarkers of SVD in predicting clinical outcomes. Meanwhile, much effort must be put into reaching standardisation of WMH assessment. Optimal characterisation of brain vascular health in congenital heart disease would enable physicians to adopt prompt strategies to prevent the risk of cognitive deterioration in this category of patients. Full translation of research findings into clinical practice would then be achieved.