Toward Quantifying Transitional Blood Flow in Patients

The dynamics of blood flow in healthy and diseased vessels has long interested clinicians, physicists, engineers, and vascular biologists. Of particular note is the transition of blood flow to unstable and potentially turbulent flow regimes, which draws attention from all disciplines. A solid understanding of how flow transitions in the model problem of “wet water running through a pipe” has been obtained [1]. However, the factors driving instability in blood flow remain elusive. Potential factors include flow pulsatility [2], vessel wall motion [3], the complex rheological properties imparted by red blood cells [4], as well as the wide variety in vessel geometries in healthy and diseased states. In-vitro [5, 6] and in-silico [7‒9] approaches excel at isolating these factors but fall short in determining when blood flow transitions occur in patients. Ultimately, the most suitable place to study when instabilities in blood flow occur is in human vessels themselves.The primary challenge in studying blood flow instabilities is the need for measurement techniques with a high temporal resolution to capture instantaneous cardiac cycles. This restricts the use of traditional phase-contrast magnetic resonance imaging of blood flow, in which interleaved sampling filters out transient fluctuations that can vary from cycle to cycle. To address this, alternative MRI sequences have been developed that detect intravoxel signal loss caused by incoherent turbulent velocity fluctuations. These sequences have evolved into clinically feasible scanning strategies [10], with ongoing refinements to improve the accuracy of turbulent energy quantification [11]. While these advancements are promising and could serve as reference modality for quantifying turbulent characteristics in-vivo, their high cost and complexity may limit widespread clinical application in large patient cohorts.Ultrasound techniques are inherently faster and more scalable than MRI but, in current clinical practice, are limited to 2D imaging of blood flow. Doppler-based techniques are furthermore sensitive to assumptions about flow direction. Transitional and turbulent flows are characterized by three-dimensional velocity structures, which can meaningfully violate these assumptions. In turbulent flow, rapid temporal or spatial variations lead to high variance in encountered Doppler shifts – a feature that has been exploited early in the development of color Doppler by adding green hues at these samples [12]. Thus, the increased variability or incoherence in the velocity signal during turbulent blood flow serves as a valuable source of information that can be leveraged by both MRI and ultrasound techniques.In a patient study, Cerminati et al. [13] capitalized on color Doppler’s ability to localize sites of high velocity variance by quantifying the structure of green hues. By employing methods from fractal analysis, the authors developed an algorithm to measure oscillations in the structure of green hues, thereby quantifying flow complexity as an indicator of turbulence. Counterintuitively, the study revealed higher flow complexity in 7 patients with no carotid stenosis compared to 37 patients with <50% stenosis. The advantage of the method is its scalability, as it can be readily applied to color Doppler data and it demonstrated decent interobserver variability.A central question remains: does this metric truly quantify flow instabilities and, if so, what specific aspects does it capture? It is possible that a <50% stenosis structures the flow – as suggested by the authors – leading to reduced flow complexity compared to non-stenosed cases in the carotid bifurcation, an area long recognized for its complex flows [14]. Alternatively, the non-stenosed group, which was on average 15-year younger, may have exhibited greater wall pulsatility, thereby increasing oscillations in flow structures. To determine if the method captures transitional and turbulent flow aspects, it would be beneficial to apply it in in-vitro experiments where the velocity field and turbulent characteristics can be quantified in detail through experimental and numerical techniques. Additionally, evaluating the metric in patients with high-grade (>75%) carotid artery stenosis would be valuable, as the turbulent breakdown of the post-stenotic jet [9] should be associated with oscillations in flow structures. These validation steps are essential to confirm that the developed metric captures blood flow’s transition toward instability.The author has no conflicts of interest to declare.The author has not received any financial support.L.V.: writing and review of the final article.