Feasibility and precision of transcutaneous very-high resolution ultrasound for quantification of arterial structures in human neonates – Comparison with conventional high resolution vascular ultrasound imaging

Johnny K.M., Sundholm, Rasmus F.W., Olander, Tiina H., Ojala, Sture, Andersson, Taisto, Sarkola

Atherosclerosis |

Background: Non-invasive transcutaneous very-high resolution ultrasound (VHRU, 25-55MHz) has recently been developed to quantify superficial vascular structures in humans. The performance of the method has yet not been evaluated invivo in neonates. The aim of the study was to compare VHRU with conventional high-resolution ultrasound (HRU, 7-12MHz), and to assess the feasibility and precision of VHRU in this population. Methods: 150 images from central elastic (common carotid, CCA) and peripheral muscular (brachial, BA; femoral, FA) arteries were obtained in 25 neonates of different gestational ages (range 33+0 to 41+5 gestational weeks) and weights (range 1570-4950g) with VHRU, and the use of HRU for comparison assessed in five. Results: Images were captured from CCAs with 35MHz, FAs using 35 and 55MHz, and BAs using 55MHz. 12MHz was unable to assess FAs and BAs, and the CCA IMT was grossly overestimated compared with 35-55MHz. IMTs of the smallest BAs and FAs were beyond the axial resolution of VHRU (<0.05mm), thus immeasurable. For VHRU, the intra-, inter- and test-retest coefficients of variation (CV) were for LDs (range 1.44-2.62mm, CVs between 1.6 and 4.8%), IMATs (range 0.141-0.161mm, CVs between 8.8 and 19.9%), and IMTs (range 0.062-0.165mm, CVs between 12.8 and 24.8%) for the different arteries. Conclusion: VHRU is feasible, accurate and precise in the assessment of superficial proximal conduit arteries but unable to assess the abdominal aorta in human neonates HRU-derived neonatal conduit arterial wall layer thicknesses are below the ultrasound axial resolution.