1 Division of Experimental Medicine, Department of Medicine, McGill University.
2 Cardiovascular Health Across the Lifespan (CHAL) Program, Research Institute of the McGill University Health Centre.
3 Division of Internal Medicine, Department of Medicine, McGill University Health Center.
4 Centre for Outcomes Research and Evaluation (CORE), Research Institute of the McGill University Health Centre, McGill University, Montreal, Quebec, Canada.
5 Biomedical Engineering Unit, First Department of Cardiology, Medical School, Hippokration Hospital, National and Kapodistrian University of Athens, Greece.
6 Faculty of Medicine and Dentistry, University of Alberta, Edmonton, Alberta.
7 Metabolism, Obesity and Nutrition Laboratory, PERFORM Centre.
8 Department of Health, Kinesiology, and Applied Physiology, Concordia University, Montreal, Quebec, Canada.

Introduction: Accurate comparisons of carotid--femoral pulse wave velocity (cfPWV) within and across studies require standardized procedures. Guidelines suggest reporting the average of at least two cfPWV measurements; if the difference exceeds 0.5 m/s, a third measurement should be taken, and the median reported. Another method involves repeating measurements until two values are within 0.5 m/s. However, in many studies, duplicate measurements are averaged irrespective of the difference between readings. We evaluated the impact of these methods on the reported cfPWV value.

Methods: Measurements of cfPWV (SphygmoCor) from five studies included individuals spanning a wide age range, with or without comorbid conditions, and pregnant women. In participants with at least three high-quality measurements, differences between the median value (MED) and the average of the first two cfPWV measurements (AVG1) and the average of two cfPWV measurements within 0.5 m/s (AVG2) were evaluated using paired t-tests and Bland--Altman plots.

Results: Participants' mean age was 50 ± 14 years and BMI was 28.0 ± 5.5 kg/m (N = 306, 79% women). The overall mean difference was -0.10 m/s (95% CI 0.17 to -0.04) between MED and AVG1, and 0.11 m/s (95% CI 0.05--0.17) between MED and AVG2. The absolute difference exceeded 0.5 m/s in 34% (MED-AVG1) and 22% (MED-AVG2) of participants, and 1 m/s in 8% of participants (both MED-AVG1 and MED-AVG2). Scatter around the bias line increased with higher mean cfPWV values.

Conclusion: Although the overall mean difference in cfPWV between protocols was not clinically relevant, large variation led to absolute differences exceeding 0.5 m/s in a large proportion of participants.