We aimed to develop a method of evaluating arterial stiffness using oscillometric blood pressure measurement. The major new findings in this study were that a curve relationship between cuff pressure and arterial volume could be derived from blood pressure pulse obtained from oscillometric blood pressure measurement, and that after fitting an equation to the curve, a numerical coefficient of the equation (arterial stiffness index) was related to PWV(baPWV and cfPWV) and carotid arterial compliance. Furthermore, a stepwise multiple regression analysis demonstrated that baPWV and carotid arterial compliance were the independent determinants of API, and that API was the independent determinant of baPWV and carotid arterial compliance. These results suggest that arterial stiffness can be evaluated using our method based on oscillometric blood pressure measurement.
Only few methods have so far been proposed to assess arterial stiffness using oscillometric blood pressure measurement [23
]. Liu et al. proposed a method for evaluating brachial arterial compliance using the oscillometric pulse [23
]. Their method is based on oscillometric blood pressure measurement, but requires a record of not only the cuff pressure with a pressure transducer, but also the cuff volume with an air flow meter. The method examined by Sato et al. assesses arterial stiffness with evaluating the shape of the oscillometric envelope [24
]. This method requires only cuff pressure recording, but the calculated index of arterial stiffness varies so that the measurement was repeated five times and the maximum and minimum values were excluded [24
]. In the present study, we proposed a method that can assess arterial stiffness with recording only cuff pressure. Additionally, the coefficient of variation of replicate measurements at the experimental day for API was 7.5 ± 6%, which is comparable to that for cfPWV (7 - 8%) previously reported [19
]. Furthermore, the day-to-day coefficient of variation for API in our pilot study was 6.0 ± 1.1%, whereas that for cfPWV and carotid arterial compliance in the previous studies were 5 - 9% [4
]. Taken together, our method can assess arterial stiffness with recording only cuff pressure and the calculated arterial stiffness index (API) has reproducibility that is comparable to cfPWV and carotid arterial compliance.
It is possible that the slope of the pressure-volume curve in this study correlates with brachial artery stiffness. When blood pressure is measured oscillometrically, decreasing cuff pressure increases transmural pressure of the brachial artery wall, causing the brachial artery to distend and arterial volume to increase. Vessel distension corresponding to transmural pressure would be greater in a compliant than in a stiff brachial artery. Hence, the slope of the pressure-volume curve between cuff pressure and arterial volume would be steeper for those with a compliant brachial artery. This concept is in agreement with previous studies that have assessed relationships between changes in intra-arterial pressure and in corresponding vessel diameter using isolated arteries from humans and animals [11
]. For example, the slope of the pressure-diameter curve of the atherosclerotic iliac artery is gentle and the calculated elastic modulus of the arterial wall increases in dogs fed with a high-cholesterol diet for 12 months [15
]. Furthermore, a computer simulation that described the curve between transmural pressure and arterial volume during deflation of a cuff wrapped around the upper arm also found a steeper slope in compliant than in stiff brachial arteries [10
]. According to the present and previous findings, the slope of the curve between the cuff pressure and arterial volume seems to vary depending on the stiffness (compliance) of the brachial artery. Thus, the arterial stiffness index (API) developed herein would reflect brachial arterial stiffness.
Although the API would reflect brachial arterial stiffness, it correlated with cfPWV, an arterial stiffness index of central arteries such as the aorta. One interpretation of this finding is that individuals with stiff (or compliant) central arteries also have stiff (or compliant) peripheral arteries. Arterial stiffness of the central arteries increases with advancing age [4
], whereas the effect of age on the peripheral arteries is still controversial. However, API that is derived from brachial artery pulses was negatively correlated with age in the present study (Figure ). In agreement with this finding, the large peripheral arteries of the arms and legs as well as the aorta stiffen with age [27
], although arterial stiffness with age is relatively modest in the peripheral arteries compared with the central arteries [27
]. Different from the modest effects of age, regular endurance exercise in humans increases both femoral and carotid arterial compliance [29
] and a high cholesterol diet in dogs increases iliac and carotid arterial stiffness [15
]. It was noted that the effects of the exercise and high cholesterol diet on compliance/stiffness of femoral and iliac artery were greater than that of carotid artery [15
]. Collectively, aging and lifestyle such as exercise and diet alters peripheral and central arterial stiffness in parallel, although the effects of age on peripheral artery might be modest, and thus API might be correlated with cfPWV.
API was correlated with PWV and carotid arterial compliance but the correlation coefficients were relatively low. We consider that the low correlation coefficients are probably due to the methodology differences rather than the poor validity of API. Indeed, the correlation coefficients between carotid arterial compliance and baPWV or cfPWV were also relatively low in the present study (r = -0.22; r = -0.14), despite the accepted methods clinically and experimentally. One explanation for the law correlations among methods is that each method does not entirely assess the same artery. Additionally, each approach may partly evaluate the same property of the arterial wall, but may also assess the different arterial properties. These might affect the correlation coefficients among API and PWV or carotid arterial compliance.
Our developed method has important clinical implications. We developed an arterial stiffness index based on oscillometric blood pressure measurements. If the algorithm to calculate the index was added to an oscillometric device, blood pressure and arterial stiffness could be simply and simultaneously measured. This would enable the early detection of arterial stiffness and thus contribute to the prevention of cardiovascular disease. Additionally, the API correlated with cfPWV, which is a stiffness index of the central artery. To assess central arterial stiffness is clinically important because it is a predictor of cardiovascular mortality [30
]. The correlation between cfPWV and API suggests that the developed index could be used as a tool to screen central arterial stiffness. Furthermore, our developed methodology would be useful for daily control of arterial stiffness. Endurance exercise training for 2-3 months improved arterial stiffness [4
], suggesting that arterial stiffness could be controlled in daily life. The developed method would match this demand because it is simple and easy for use.
The potential limitations in this study should be discussed. Firstly, we evaluated arterial volume indirectly via the oscillometric cuff pressure. Because, the muscles and the fat exist between the brachial blood vessel and the cuff, the muscle or fat size may affect on our estimated arterial volume, thereby API. Indeed, a stepwise multiple regression analysis revealed that the circumference of the upper arm was an independent determinant of API. This should be taken into account to use API. Secondly, we did not evaluate arterial stiffness among individuals with diseases that are related to this condition such as diabetes. Further studies are needed to examine whether our method can detect arterial stiffness in such patients.