Introduction
Ultrasonographic (US) measurements of intima-media thickness (IMT), particularly of the carotid arteries, have been used as a cardiovascular risk assessment tool. Variations in manual and semiautomatic measurements were investigated in an animal model—the aorta of 10 rabbits with dyslipedemia.
Methods
IMT was measured in B-mode images with 256 shades of gray using a 10 MHz probe by two independent investigators: (a) three same-place manual measurements were averaged in five distinct locations of the aorta of each rabbit for a total of 50 measurements per investigator (subgroups M1 and M2); (b) experimental software allowed for alterations in blood-intima and media-adventitia brightness line definition for IMT detection; observer 1 used brightness 50–200, adaptable to all images (SA1); observer 2 started with 10–60—the usual for human carotid measurements in this package—and increased values until IMT lines were visually acceptable; IMT was measured at optimized values for each image (SA2). Mean values were compared by using the paired t-test available in Excel.
Results
SA1 measurements, 0.379 ± 0.112 (SD) mm, were significantly smaller than SA2, 0.403 ± 0.122 mm, M1, 0.404 ± 0.149 mm, and M2, 0.405 ± 0.140 mm (<0.001 for all three SA1 versus SA2, M1 and M2 comparisons).
Conclusion/Discussion
Distinct automatic technologies, including (a) selection of blood-intima and media-adventitia brightness interfaces and (b) signal processing to “improve” line images, may alter IMT measurements significantly. Assumed population variations may be affected by technology used. IMT gender and age tables may be applicable only if the same measurement technology was used consistently.