Sunday, 19 May 2013

Niblet by Andrew Cummings

Getting into minutiae in a big way….

You can know the name of a bird in all the languages of the world, but when you're finished, you'll know absolutely nothing whatever about the bird...”
 Richard Feynman

In our investigation centric age you are likely someday to find yourself nodding blandly at an MRI head report with the words “age related changes with SVD only” hanging loosely in the opening sentence.  For all our sense of familiarity with small vessel disease (SVD) its apparent unobtrusiveness disguises a world of complexity. 

The putative role for SVD in neurological, psychiatric and cognitive disease burden is huge.  It is responsible for around a fifth of all strokes, doubles the overall risk and contributes to around half of all dementias. In this months Lancet Neurology, Joanna Wardlaw produced a well thought out review which goes into detail about the who’s who of this disease.

But lets go back a step.  What actually is SVD?  On pathological sections, typical T1-, T2-weighted imaging and FLAIR scans, SVD appears as a combination of small subcortical hemorrhages and infarctions, their resultant lacunes (CSF filled cavities) and brain atrophy.  With the advent of more advanced forms of imaging such as diffusion tensor imaging and magnetization ratio, the more indolent effects of SVD; altered myelination, focal thinning of cortical grey matter and even disruption to axonal transport have been observed.  This is all well and good, but the key question is what is the actual cause of these phenomenon?

At a histopathological level SVD appears as a diffuse, intrinsic disease of small arterioles with a characteristic appearance of damage from inflammatory cell infiltration into the vessel wall and perivasculature.  Hypertension, vasospasm and microatheroma have all been put forward as possible mechanisms for this. 

But there’s a difficulty…

The bare bones of it is that while SVD and cortical strokes are globally associated with vascular risk factors, in studies of normotensive patients with SVD verified at autopsy a majority showed limited vascular risk factors.  While hypertension is known associated with the development of white matter hyperintensities (another radiological sign of SVD) the mechanism by which these lead to vascular damage is convoluted at best.  There is a suggestion, that part of the observed association we see between SVD and hypertension may actually be due in part to co-localisation of genetic loci for susceptibility to both these conditions.

So if hypertension isn’t the only answer, what else might be?    Wardlaw and colleagues noticed that in many pathological specimens of lacunar infarcts, a central perforating arteriole with a thickened wall runs through the centre of the infarct rather than proximal to it.  This suggests that rather than occlusion and resultant hypoxia occurring downstream of an atheromatous lesions (as we are familiar with), something else is happening directly around the affected vessel that leads to the eventual infarction.  That something, the authors suggest, is endothelial dysfunction. 

The cells of the cerebrovascular endothelium are most closely “packed” at the level of capillaries with widening gap junctions in arterioles and venules.  With advancing age, inflammation and oxidative stress this barrier becomes more permeable.  Interestingly, in animal models of SVD, short exposure to salt (a known risk factor) is enough to cause a worsened degree of the latter two of these alongside small vessel pathology.  The authors suggest that in areas of the brain which have impaired endothelial function, a leakage of plasma contents and migration of inflammatory cells ensues.  This leads to distortion of the arteriolar lumen eventually leading to thrombosis. Perivascular tissue damage also takes place, resulting in the white matter demyelination and dilatation of perivascular spaces observed as hyperintensities on MRI. 

By Andrew Cummings (Academic F2, Barts Health)

1 comment:

  1. Great post. I like the 'leaky vessel' hypothesis.

    There's an interview with Joanna Wardlaw in the latest Lancet Neurology podcast, for anyone who likes that sort of thing (I highly recommend the whole podcast series - always full of interesting discussion!):