Cerebral haemodynamics in patients with glutaryl-coenzyme A dehydrogenase deficiency

In glutaric aciduria type 1, glutaryl-coenzyme A and its derivatives are produced from intracerebral lysine and entrapped at high concentrations within the brain, where they interfere with energy metabolism. Biochemical toxicity is thought to trigger stroke-like striatal degeneration in susceptible children under 2 years of age. Here, we explore vascular derangements that might also contribute to brain damage. We studied injured and non-injured Amish glutaric aciduria type 1 patients using magnetic resonance imaging (n = 26), transcranial Doppler ultrasound (n = 35) and perfusion computed tomography (n = 6). All glutaric aciduria type 1 patients had wide middle cerebral, internal carotid and basilar arteries. In non-injured patients, middle cerebral artery velocities were 18-26% below control values throughout late infancy and early childhood, whereas brain-injured children had an early velocity peak (18 months) and low values thereafter. Perfusion scans from six patients showed that tissue blood flow did not undergo a normal developmental surge. We observed four different perfusion patterns. (i) Three children (two non-injured) had low cerebral blood flow, prolonged mean transit time, elevated cerebral blood volume and high mean transit time/cerebral blood flow and cerebral blood volume/cerebral blood flow ratios. This pattern optimizes substrate extraction at any given flow rate but indicates low perfusion pressure and limited autoregulatory reserve. (ii) Ten hours after the onset of striatal necrosis in an 8-month-old infant, mean transit time and cerebral blood volume were low relative to cerebral blood flow, which varied markedly from region to region. This pattern indicates disturbed autoregulation, regional perfusion pressure gradients, or redistribution of flow from functional capillaries to non-exchanging vessels. (iii) In an infant with atrophic putaminal lesions, striatal flow was normal but mean transit time and cerebral blood volume were low, consistent with perfusion in excess of metabolic demand. (iv) Finally, a brain-injured adult with glutaric aciduria type 1 had regional perfusion values within the normal range, but the putamina, which normally have the highest regional perfusion, had cerebral blood flow values 24% below cortical grey matter. Although metabolic toxicity appears central to the pathophysiology of striatal necrosis, cerebrovascular changes probably also contribute to the process. These changes may be the primary cause of expanded cerebrospinal fluid volume in newborns, intracranial and retinal haemorrhages in infants and interstitial white matter oedema in children and adults. This pilot study suggests important new areas for clinical investigation.