Mark N. Rubin, MD

Nael K, Trouard TP, Lafleur SR, Krupinski EA, Salamon N, and Kidwell CS. White Matter Ischemic Changes in Hyperacute Ischemic Stroke: Voxel-Based Analysis Using Diffusion Tensor Imaging and MR Perfusion. Stroke. 2014

Multiparametric “physiologic” acute stroke imaging (e.g., CT or MR perfusion) is all the rage in academe and clinical practice alike. This is because these imaging modalities provide what is thought to be a representation of microstructural changes in the brain – as in what happens when membrane pumps stop doing their thing – and a way to differentiate how much of the brain can be salvaged as opposed to a sense that the proverbial horses have already left the barn, thus the door need not be slammed. These data, in turn, inform early management.


Some University of Arizona investigators sought to describe another physiologic imaging parameter in acute stroke: fractional anisotropy (FA) in affected white matter tracts. So, if our familiar diffusion-weighted MRI sequences for stroke (e.g., DWI) give us information on diffusion restriction (as a biomarker of neuronal injury) in a region of brain, testing FA in white matter tracts (e.g., DTI sequence) tells us about diffusion restriction along the scanned extent of the white matter tract. The authors of this manuscript hypothesized that changes in FA may be a useful surrogate for loss of structural and/or functional integrity in the hyperacute phase of ischemic stroke, and compared this to other more recognizable sequences such as ADC & Tmax with perfusion. An ADC threshold was set for infarct vs hypoperfusion.

This was a small, exploratory study which recruited 21 patients (14 male) within 6 hours of acute ischemic stroke and a demonstrable “penumbra” (e.g., diffusion-perfusion mismatch). The sequences of interest, namely DTI, ADC and perfusion-based Tmax, were co-registered and quantified (voxel-based, third-party software) and compared to the homologous region on the contralateral side as an “internal control.”

The findings based on the primary hypothesis were interesting, and have some overlap with more recognizable imaging variables that suggest potential clinical utility. As expected, there was a significant difference in FA between normal, hypoperfused and infarcted white matter. What was interesting was that the pathologic states varied in opposite directions; FA increased in the hypoperfused white matter and decreased in the infarcted white matter. These findings are consistent with previous studies suggesting that cytotoxic edemia in white matter may restrict diffusion but not greatly affect the cellular architecture, whereas decreasing FA with infarct suggests uniform restriction of diffusion in the region which would be expected with frank injury.

Multiparametric imaging, although clinically mainstream and deserving of the many research dollars invested, is yet imperfect. Studies like these, exploring other physiologic parameters that may serve as that elusive “should I or shouldn’t I?!” biomarker for acute stroke treatment outside of the typical clinical indications, are a way to improve upon diagnosis and management of our acute stroke patients.