Can Arterial Stiffness Be Measured on a Bathroom Scale?

The history of technological development is full of examples in which the often unusual coupling of two separate principles into a single device led to breakthrough technological advancements. Around 1450 in Strasbourg, Gutenberg combined the flexibility of a coin punch with the power of a wine press and invented the printer. Probably no one at that time would expect how much this invention contributed to spread the knowledge throughout the entire world in later centuries. In the field of hypertension, the combination of principles of mercury sphygmomanometry, invented by Riva Rocci in 1890,1 with the technique of auscultation of blood flow sounds described by Korotkoff in 1905,2 enabled for the first time the measurement of systolic and diastolic brachial blood pressure values. The outstanding and unpredictable success associated with the possibility to obtain measures of blood pressure is well acknowledged. Needless to say that for any given successful technological combination, many others will have turned out as unfavorable combinations. In the current issue, Campo et al. present the results of a novel device in which the principles of ballistocardiography and local impedance plethysmography are combined together to obtain a measure of aortic-to-leg pulse wave velocity (al-PWV) 3. Curiously, these technologies have been assembled on a bathroom scale (the Whithings scale), an object conceived for common daily usage. The ballistocardiography is the detection of the rhythmic body mass displacements induced by forces associated with cardiac contraction and blood flow accelerations. The principle is based on the Newton’s third law: as blood is ejected by the left ventricle, an upward action vector is generated, and the reaction impels the body toward the feet. Pioneering works dated back toward 1877,4 but the advent of other contemporary technologies cooled down the first enthusiasms. Typically, ballistocardiography was obtained with the subject lied on a suspended table and very small movements are recorded on a paper. The use of electromechanical film sensors and digital signal processing enabled the evaluation of ballistocardiography in the upright position.5 On a typical ballistocardiogram, all the phases of cardiac cycle are represented, and the beginning of systole is marked by a head-ward wave (H-wave) which peaks approximately with aortic valve opening6. Ballistocardiography is highly sensitive to external noises and body movements. The principle of electrical impedance plethysmography is related to the Ohm’s law, which states that current between 2 points is directly proportional to the voltage across these points, and inversely proportional to resistance. When a differential electrical current is applied between 2 points of body surface, changes in volume blood pulse generated by arterial blood inflow and venous outflow originate phasic changes in electrical impedance, since electrical impedance is related to ionic conductivity of blood fluids. The technology of electrical impedance plethysmography has been extensively applied to measure cardiac output7 and to detect abnormalities in venous limb blood flow.8 When changes in local impedance are represented in a time-dependent manner, the wave changes in electrical impedance in a distal segment of the body correspond to the arrival of blood flow wave. The combination of ballistocardiography with impedance pletysmography allows to detect the time lag between each heart contraction and the arrival of the pressure wave to feet soles. In the Whithings Scale device, this measure is then converted into al-PWV using a travel distance estimated from the height of the subject. In the present study, the al-PWV assessed through the Whithings scale was compared with carotid-femoral PWV (cf-PWV) measured by the Sphygmocor. In the first part of the study, the aortic-to-leg pulse transit time was derived by the integration of ballistocardiography and local impedance pletysmography in 106 subjects, and data were optimized with height-derived measures in order to regress al-PWV from aortic-to-leg pulse transit time. In the second part, a formal comparison between al-PWV and cf-PWV was performed. Results showed a moderate degree of correlation between al-PWV and cf-PWV measures (Spearman’s rho = 0.70, P < 0.001), with an absolute average difference of 0.25 m/s and a SD of 1.39 +m/s. In the third part of the study, the authors evaluated the contribution of main determinants of arterial stiffness to both al-PWV and cf-PWV. Interestingly, age showed a closer age-dependency with al-PWV (beta-coefficient 0.55) than with cf-PWV (beta-coefficient 0.38). Taken together, are these results sufficient to conclude that al-PWV could parallel the measurement of cf-PWV in clinical practice? What are similarities and differences between these 2 measures? The cf-PWV is usually conceived as the noninvasive gold standard method to assess aortic stiffness.9 Cf-PWV is not only prognostically relevant, but also more informative that the combination of traditional CV risk factors, and improves the model fit for CV risk prediction beyond traditional risk factors, including office brachial BP.10 Studies aiming at employing cf-PWV measures as a guide to antihypertensive therapy11 are ongoing, and results are warmly awaited. There is also mounting interest on devices measuring the speed of pressure wave transmission along different segments of the vascular tree, in part because these measurements are usually performed in a less intrusive and operator-independent way. Both brachial-ankle PWV12 or the cardio-ankle vascular index (CAVI), which is based on heart-ankle PWV,13 are measured on longer and heterogeneous arterial segments, including the aorta and peripheral muscular arteries. These measures are becoming increasingly popular and showed an independent association with markers of organ damage14,15 and worse CV outcome, especially in Asian populations.16,17 A number of studies previously showed that brachial-ankle PWV is strongly associated with cf-PWV,18,19 thus reflecting a higher sensitivity to changes in stiffness of central arteries rather than peripheral muscular arterial segments. Because stiffness of distinct arterial segments varies differently in response to aging and extrinsic stimuli, such as hypertension and other classical CV risk factors,20 sources of discrepancy may be identified in specific clinical conditions specifically affecting more peripheral arterial segments of the lower limbs than the aorta, as it occurs in DM and PAD.21 Therefore, PWV measured on different arterial segments ultimately provide different clinical information, and the demonstration of a closer association between 2 PWV measurements cannot be used as an essay of interchangeability. As compared to carotid-femoral arterial path length, the aortic-to-leg arterial segment includes the ascending aorta and the tract between the femoral and the ankles, whereas the segment from carotid pulse to aortic arch is omitted. The finding that al-PWV and cf-PWV are rather similar in terms absolute values, as reported in the present article, is quite unexpected, at least from a theoretical perspective. Interestingly, a study from Sugawara22 provided magnetic resonance imaging-derived path length and related PWV values of heart-ankle PWV, brachial-ankle PWV, and cf-PWV. As it can be derived from published data, heart-ankle PWV (which should more closely reflect al-PWV) is univariately higher than cf-PWV (Table 1) in both sexes and among different age classes. This suggest that the effective path length used for al-PWV calculation in the Whithings scale could be underestimated. Comparison between carotid-femoral pulse wave velocity (cf-PWV) and heart-ankle pulse wave velocity (ha-PWV) obtained from MRI-based measurement of effective arterial path length (from Sugawara et al.19 ) Abbreviation: MRI, magnetic resonance imaging. Comparison between carotid-femoral pulse wave velocity (cf-PWV) and heart-ankle pulse wave velocity (ha-PWV) obtained from MRI-based measurement of effective arterial path length (from Sugawara et al.19 ) Abbreviation: MRI, magnetic resonance imaging. A number of studies also showed that peripheral muscular arteries usually exhibit less remarkable changes with aging and diseases than aorta and central arteries.23,24 To the best of our knowledge, there are no studies comparing the age-dependency of cf-PWV and heart-ankle PWV. In our view, the closer age-dependency of al-PWV, as compared to cf-PWV, as it was reported in the present study, could have a different explanation. Since al-PWV is measured in the upright position, the arterial path is exposed by a higher hydrostatic BP gradient. Arterial stiffness is not a static measure, but undergoes dynamic increases for any given increment in distending pressure.25 Typically, this pressure dependence of arterial stiffness is described by a nonlinear (usually exponential), rather than linear, relationship26, and an increased curvilinearity is the result of the impact of age and other CV risk factors. Therefore, with advancing age, changes in distending pressure on the aortic-to-leg arterial path when assuming the upright position may be associated with increased values of al-PWV, which can only in part be estimated from BP hydrostatic variations. This was clearly shown in previous studies exploring the relationship between acute changes in pressure gradient and arterial stiffness-related parameters, both in normal subjects and in subjects with pronounced orthostatic BP changes due to autonomic failure. Results demonstrated that changes in PWV could be only in part predicted by BP differences recorded from supine to standing position27. Data about the clinical usefulness of PWV estimated in the aortic-to-leg arterial path are limited and, more importantly, this measure lacks of formal invasive validation studies. In this view, the al-PWV evaluated with the Whithings scale may be conceived as a novel opportunity to assess the arterial stiffness in the upright position and along a relatively unexplored arterial segment that ultimately less has to deal with the traditional and well-established cf-PWV assessment. Will the combined technology of the Whithings scale have the same success of the Guttemberg’s printer or the Riva-Rocci sphygmomanometer? Further clinical studies will tell us. All authors had access to the data and a role in writing the manuscript.