Mathieu Collette

Modeling and interpretation of the bioelectrical impedance signal for the determination of the local arterial stiffness

PURPOSE Stiffness of the large arteries (e.g., aorta) plays an important role in the pathogenesis of cardiovascular diseases. To date, the reference method for the determination of regional arterial stiffness is the measurement of the carotid-femoral pulse wave velocity (PWV) by tonometric techniques. However, this method suffers from several drawbacks and it remains limited in clinical routine. METHODS In the present study, the authors propose a new method based on the analysis of bioelectrical impedance (BI) signals for the determination of the local arterial stiffness. They show, from a theoretical model, a novel interpretation of the BI signals and they establish the relationship between the variations in the BI signal and the kinetic energy of the blood flow in large arteries. From this model, BI signals are simulated in the thigh and compared to experimental BI data. Finally, from the model, they propose a new index (Ira) related to the properties of the large artery for the determination of the local arterial stiffness. RESULTS The results show a good correlation between the simulated and the experimental BI signals. The same variations for both of them with different characteristics for rigid and elastic arteries can be observed. The measurement of the Ira index on 20 subjects at rest (mean age of 44 +/- 16 yr) for the determination of the local aortic stiffness presents a significant correlation with the PWV reference method (R2=0.77; P<0.0001 with the Spearman correlation coefficient and Ira= 4.25*PWV+23.54). CONCLUSIONS All the results suggest that the theoretical model and the new index could give a reliable estimate of local arterial stiffness.

Rigidité artérielle mesurée par pOpmètre® chez les patients à risque cardiovasculaire, lien aux plaques d’athérome carotidien

PURPOSE Aortic stiffness is a functional and structural consequence of ageing and arteriosclerosis. Regional arterial stiffness can be easily evaluated using pOpmetre(®) (Axelife SAS, France). This new technique assesses the pulse wave transit time (TT) between the finger (TTf) and the toe (TTt). Based on height chart, regional pulse wave velocity (PWV) between the toe and the finger can be estimated (PWVtf). pOpscore(®) index is also calculated as the ratio between PWVtoe and PWVfinger and can be considered as a peripheral vascular stiffness index. The aim of the study was to evaluate the relationship between pOpmetre(®) indices and the presence of carotid plaques in a population with cardiovascular risk factors. METHODS In 77 consecutive patients recruited for a vascular screening for atherosclerosis (46 men aged 54 ± 2 years; 31 women aged 49 ± 3 years; ns), the difference between TTt and TTf (called Dt-f), the regional pulse wave velocity between the toe and the finger (PWVtf = constant × height/Dt-fm/s) and pOpscore(®) were measured by pOpmetre(®). Presence of carotid plaques was assessed using ultrasound imaging. The local aortic stiffness (AoStiff) was evaluated by the Physioflow(®) system. RESULTS No difference was found between patients with or without carotid plaques (n=25 versus 52) for Ankle-Brachial Pressure Index (ABPI: 1.15 ± 0.04 versus 1.12 ± 0.03), nor for diastolic or systolic blood pressure (87 ± 3 versus 82 ± 2; 137 ± 3 versus 132 ± 2 mmHg). The first group was older than the second (59 ± 2 versus 49 ± 2 years, P<0.002) with a larger intimae media thickness (0.69 ± 0.02 versus 0.63 ± 0.01 mm, P<0.004), a higher AoStiff (10.4 ± 0.7 versus 8.2 ± 0.5m/s, P<0.02), and PWVtf (14.3 ± 1.0 versus 10.7 ± 0.7 m/s, P<0.004) and a shorter Dt-f (57.9 ± 5.1 versus 73.5 ± 3.5 ms, P<0.01). PWVtf (r(2)=0.49, P<0.0001) and Dt-f (r(2)=0.54, P<0.0001) correlated with age. A significant difference in pOpscore(®) index was observed between both groups (1.51 ± 0.3 versus 1.41 ± 0.2, P<0.006). CONCLUSION Our results show a significant arterial stiffness indices measured by pOpmetre(®) in patients with and without carotid plaques.

Assessment of aortic stiffness by local and regional methods.

The stiffness of large arteries has an important role in cardiovascular hemodynamics. Aortic stiffness (AoStiff) can be assessed non-invasively with regional and local methods. In this paper, we compared these two techniques for evaluating AoStiff. Our subjects comprised of 118 consecutive patients (85 men, mean age: 49±14 years). We evaluated regional AoStiff with carotid-femoral pulse wave velocity (PWV) measured with a tonometric technique and by bioelectrical impedance (BI) wave velocity (IWV). The local AoStiff was calculated from BI signals recorded at the chest. We used glyceryl trinitrate (GTN) to test the effect of peripheral vasodilatation on both methods in a subgroup of 52 patients (37 men, mean age: 52±11 years). We found a significant correlation between IWV and PWV measurements (r=0.88, P<0.0001) as well as between AoStiff and PWV measurements (r=0.75, P<0.0001). GTN administration decreased mean arterial blood pressure by 4% (95% confidence interval: 2–8%, P=0.002) without significant changes in AoStiff and regional IWV. Local AoStiff is correlated with regional measurements and is not influenced by changes in arterial pressure because of systemic peripheral vasodilatation.

e-PWV: A Web Application for Assessing Online Carotid-Femoral Pulse Wave Velocity

This article presents e-PWV, a web application for the determination of Pulse Wave Velocity (PWV) running on a web platform called NooLib. e-PWV has been applied on signals recorded by an ultrasound system (\(PWV_{et}\), Aloka, Japan) and representing the arterial diameter changes in carotid and femoral sites. \(PWV_{et}\) measurements were compared to PWV recorded by a tonometric technique (\(PWV_{pp}\), PulsePen, Italy). The study was conducted on 120 patients. We found an excellent correlation of r = 0.95 between \(PWV_{et}\) et \(PWV_{pp}\) (\(P<0.0001\); 95% confidence interval of 0.91–0.96; \(PWV_{et}=0.88\times PWV_{pp} + 0.57\)). We observed a small offset of \(-0.33\) ms\(^{-1}\) on the Bland-Altman plot with a limit of agreement from \(-2.21\) to 1.54 ms\(^{-1}\). Our results suggest that e-PWV application can produce online a reliable marker of the regional aortic stiffness using an echotracking system.