Brain volumetric, regional cortical thickness and radiographic findings in adults with cyanotic congenital heart disease.

BACKGROUND: Chronic cyanosis in adults with congenital heart disease (CHD) may cause structural brain changes that could contribute to impaired neurological functioning. The extent of these changes has not been adequately characterized. HYPOTHESIS: We hypothesized that adults with cyanotic CHD would have widespread changes including abnormal brain volumetric measures, decreased cortical thickness and an increased burden of small and large vessel ischemic changes. METHODS: Ten adults with chronic cyanosis from CHD (40 ± 4 years) and mean oxygen saturations of 82 ± 2% were investigated using quantitative MRI. Hematological and biochemical parameters were also assessed. All subjects were free from major physical or intellectual impairment. Brain volumetric results were compared with randomly selected age- and sex-matched controls from our database of normal subjects. RESULTS: Five of 10 cyanotic subjects had cortical lacunar infarcts. The white matter (WM) hyperintensity burden was also abnormally high (Scheltens Scale was 8 ± 2). Quantitative MRI revealed evidence of extensive generalized WM and gray matter (GM) volumetric loss; global GM volume was reduced in cyanosed subjects (630 ± 16 vs. 696 ± 14 mL in controls, p = 0.01) as was global WM volume (471 ± 10 vs. 564 ± 18 mL, p = 0.003). Ventricular cerebrospinal fluid volume was increased (35 ± 10 vs. 26 ± 5 mL, p = 0.002). There were widespread regions of local cortical thickness reduction observed across the brain. These changes included bilateral thickness reductions in the frontal lobe including the dorsolateral prefrontal cortex and precentral gyrus, the posterior parietal lobe and the middle temporal gyrus. Sub-cortical volume changes were observed in the caudate, putamen and in the thalamus (p ≤ 0.005 for all regions). Cortical GM volume negatively correlated with brain natriuretic peptide (R = - 0.89, p = 0.009), high sensitivity C-reactive protein (R = - 0.964, p < 0.0001) and asymmetric dimethylarginine (R = - 0.75, p = 0.026) but not with oxygen saturations, packed cell volume or viscosity. CONCLUSIONS: We present the first comprehensive analysis of brain structure in adults with chronic neurocyanosis due to congenital heart disease. We demonstrate clear evidence for marked macro- and microvascular injury. Cyanotic patients show global evidence for reduced brain volume as well as specific foci of cortical thickness reduction. The GM volume loss correlated with hsCRP, BNP and ADMA suggesting that inflammation, neurohormonal activation and endothelial dysfunction may have important roles in its pathogenesis.