How to improve Doppler ultrasonography as we know it

Guest article by Dr Melissa Fletcher

Doppler ultrasonography is an ever developing area of ultrasound imaging, essential in assessing the flow of blood in an individual. It is important that we have a suitable method to measure the movement of blood that involves minimal disruption to the patient, which would ideally be both quick and non-invasive. It is for these reasons that ultrasound imaging is the perfect technique in such circumstances, with research carried out to help better understand how best to apply ultrasonography here. An example of one such study is that published by Schorer R et al this month (2016 Nov 12) in the Journal of Clinical Monitoring and Computing, ‘A feasibility study of color flow Doppler vectorization for automated blood flow monitoring’, with the abstract discussed henceforth.

The researchers focused on ultrasound imaging, citing the non-invasive aspect of this application as a reason as to why research in this area is merited (as opposed to catherterisation which is commonly used to examine blood flow, yet requires tubular insertion in the body). These studies were carried out using ‘computational methods’ looking at blood flow in real-time, with certain criteria being used to evaluate the benefits of using Doppler ultrasound for assessing blood flow. This research was carried out using an artificial replica of an artery, and criteria looked at included the speed of blood flow, the size of the vessel and the location of the ultrasound wave during examination. As these experiments were carried out using artificial means, whilst the blood flow could be identified it was revealed that the margin of error could fluctuate greatly, and this is important when determining the usefulness and reliability of using similar computing techniques for any future studies. Nevertheless, the margin of error could also be extremely low, and consequently this kind of research would be a useful tool to use in the future.

Any planned similar studies should take advantage of computational techniques as a way by which to continuously improve our usage of ultrasound imaging, with the results revealed likely to eventually better the patient-ultrasound experience in the future. As ultrasonography is suitable for use in all kinds of patients (the human kind as well as in animals!), this kind of research provides us with vital information that is sure to benefit the veterinary world sooner than we think.


Comment from Catherine Stowell

Studies like the one discussed above demonstrate the great progress researchers have made since early studies into observing and quantifying blood flow. In order to test various haemodynamic states, many studies of the 60s and 70s involved invasive procedures on animals, surgically narrowing vessels (for example, the aortas of rabbits) and occluding arteries.

Computational simulations not only avoid animal experimentation, but also offer the advantage of being able to test hypothesises over time frames not otherwise practical or feasible. The post-stenotic dilatation debate, for example, is one that remains unsettled, despite being the subject of research for several decades. Just distal to a stenotic aortic valve, it is not uncommon to find that the aorta is dilated. This is particularly common in individuals with a congenital cause to their aortic stenosis (typically a bicuspid valve), which suggests that the aortic walls may be intrinsically weak as part of the syndrome, and/or that the unusual haemodynamics created by a valve which is not tricuspid (i.e. the eccentricity and turbulence of the jets) may cause impingement of high velocity blood onto the aortic walls, thus causing dilatation. Whilst post-stenotic dilatation has been demonstrated in rabbits whose aortas were intentionally stenosed using a band, lending support to the latter theory, clearly a situation whereby a normal aorta suddenly becomes significantly narrowed over the time course of a few minutes bears little relation to real life disease progression – and thus any conclusions drawn from this must be treated with extreme caution.

It was of course impractical (not to mention cruel) to attempt to observe the evolution of a gradually narrowing aorta over time, which would of course be the natural progression in both humans and animals, but with new computational models, it becomes possible to play these scenarios out over any given time period.