Feasibility and reliability of carotid intima-media thickness measurements in nonsedated infants

J Hypertens. 2016 Nov;34(11):2227-32. doi: 10.1097/HJH.0000000000001065.


Introduction: Carotid intima-media thickness (CIMT) is a surrogate marker for atherosclerosis. It is increased in adolescents and young adults at risk for future cardiovascular disease. However, it remains unclear if it can be considered as a surrogate marker for atherosclerosis in infancy as very few studies have been performed in infants.

Objectives: Our objective was to assess the feasibility and interobserver reproducibility of CIMT measurement in nonsedated infants.

Methods: We measured CIMT in 81 infants less than 1 year of age. Repeated measurements were obtained by a second observer in 24 children. The analysis was performed with semiautomated edge detection software. Measurements with over 95% edge detection over a length of 1 cm were considered as valid. We further compared the measurements using the semiautomated method with measurements using the manual electronic caliper method in a subgroup of 10 infants.

Results: Carotid ultrasound recordings and intima-media thickness measurements were obtained in 79% of infants (n = 64). Mean CIMT of the 64 infants measured by the first observer was 0.44 mm (SD: 0.04). In the 24 participants with measurements by two observers, the mean interobserver difference was 0.001 mm (SD: 0.026). The interobserver coefficient of variation was 5.9%. CIMT measurements obtained with the manual method (mean: 0.35; range: 0.29-0.39) were slightly lower than measurements obtained with the semiautomated method (mean: 0.38; range: 0.32-0.44). Measurements with both methods were highly correlated (r: 0.87).

Conclusion: Measurement of CIMT in nonsedated infants less than 1 year of age is feasible in the majority of infants with good interobserver variability.

MeSH terms

  • Atherosclerosis
  • Carotid Intima-Media Thickness* / standards
  • Carotid Intima-Media Thickness* / statistics & numerical data
  • Feasibility Studies
  • Humans
  • Infant
  • Reproducibility of Results
  • Software*