Original contributionWithin-subject variability of flow-mediated vasodilation of the brachial artery in healthy men and women: implications for experimental studies
Introduction
Assessment of flow-mediated vasodilation (FMD) is used to evaluate endothelial function and is used as a surrogate endpoint for cardiovascular disease in observational and experimental studies. It is based on the ability of vascular endothelial cells to respond to changes in shear stress. The response consists of a release of nitric oxide, a potent vasodilator that relaxes the smooth muscle cells in the vascular wall (Joannides et al 1995), resulting in an increase in lumen diameter. FMD is generally assessed from high-resolution ultrasound (US) images of the brachial artery at baseline and at maximum dilation (Sorensen et al 1995). In healthy people, FMD is 5% to 15% of the baseline diameter but, in patients with cardiovascular disease, FMD is impaired or absent Sorensen et al 1994, Neunteufl et al 1997. An impaired FMD appears to be predictive of future events: patients with chest pain and impaired FMD were shown to be more likely to experience cardiovascular events in a 5-year period than patients with preserved FMD (Neunteufl et al 2000).
The diameter of the brachial artery can be measured reproducibly with high-resolution (7.5 to 13 MHz) US: in the literature we found coefficients of variation (CVs) (within-subject SD divided by mean baseline diameter) of 1.5% to 6%. At a mean baseline diameter of 4.0 mm, this corresponds with a within-subject SD of 0.06 to 0.24 mm. However, the reported within-subject variability of the FMD is less consistent: CVs ranging from 1.2% to 13.6% (Table) have been reported. It is possible that some of these reported CVs are, in fact, SDs, or that they represent the CVs for baseline or maximum diameter rather than for the percentual increase in diameter. Instead of CVs, some investigators report alternative measures of reproducibility that are not well-defined in the papers (Table 1). Moreover, most reproducibility studies were based on only two repeated measurements in less than 10 subjects. Therefore, we designed a study that would adequately estimate within-subject variability in FMD. This information was used to estimate sample sizes for experimental studies that compare treatment differences in flow-mediated vasodilation.
Section snippets
Subjects and methods
The Medical Ethics Committee of Wageningen University approved the study protocol. We explained aim and design of the study to the volunteers, who all signed informed consent forms.
Reading or intraobserver variance
The mean ± SD of 13 scans read twice by one observer was 4.2 ± 2.5 FMD% at the first reading and 3.3 ± 1.8 FMD% at the second reading. The correlation between the first and second reading was 0.74. The intraobserver SD was 1.27 FMD% at a mean of 3.73 FMD%, corresponding with a CV of 34%.
Within-subject variance
The number of measurements of sufficient quality ranged from 2 to 6 per subject (Table 2), with a total of 62 of 78 scans. The remaining 16 scans with low quality were not used in the analysis. The CV of the
Discussion
Within-subject variability of flow-mediated vasodilation is large: at a mean FMD of 5.60 FMD% of the baseline diameter, we found a within-subject SD of 2.82 FMD%. The corresponding CV was 50%, which is larger than those found in the literature (Table 1). However, the CVs for baseline and maximum diameter in our study were about 5%, comparable to those found by others Sinoway et al 1989, Sorensen et al 1995. Therefore, we think that previously reported CVs of FMD may either have been SD of FMD (
Acknowledgements
The authors thank the volunteers for their time and cooperation, Dr. M. van Montfort and Dr. A.A.M. Jansen for statistical advice, our sonographer, Jan Harryvan, for conducting the measurements, personnel of the Vascular Imaging Center of the Julius Centre for Health Sciences and Primare Care, Utrecht Medical Center for reading the images, and Dr. Rudy Meijer of the same center for training.
References (23)
- et al.
Non-invasive detection of endothelial dysfunction in children and adults at risk of atherosclerosis
Lancet
(1992) - et al.
Late prognostic value of flow-mediated dilation in the brachial artery of patients with chest pain
Am J Cardiol
(2000) - et al.
Systemic endothelial dysfunction is related to the extent and severity of coronary artery disease
Atherosclerosis
(1997) - et al.
Noninvasive determination of endothelium-mediated vasodilation as a screening test for coronary artery diseasePilot study to assess the predictive value in comparison with angina pectoris, exercise electrocardiography, and myocardial perfusion imaging
Am Heart J
(1999) - et al.
Occlusion cuff position is an important determinant of the time course and magnitude of human brachial artery flow-mediated dilation
Clin Sci
(2000) - et al.
Passive smoking and impaired endothelium-dependent arterial dilatation in healthy young adults
N Engl J Med
(1996) - et al.
Endothelium-dependent dilatation is impaired in young healthy subjects with a family history of premature coronary disease
Circulation
(1997) - et al.
Technical aspects of evaluating brachial artery vasodilatation using high-frequency ultrasound
Am J Physiol
(1995) - et al.
Cholesterol reduction rapidly improves endothelial function after acute coronary syndromes. The RECIFE (reduction of cholesterol in ischemia and function of the endothelium) trial
Circulation
(1999) - Hashimoto M, Akishita M, Eto M, et al. Modulation of endothelium-dependent flow-mediated dilatation of the brachial...