Attenuation of Systolic Blood Pressure and Pulse Transit ... - IEEE Xplore

346

IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, VOL. 61, NO. 2, FEBRUARY 2014

Attenuation of Systolic Blood Pressure and Pulse Transit Time Hysteresis During Exercise and Recovery in Cardiovascular Patients Qing Liu, Bryan P. Yan, Cheuk-Man Yu, Yuan-Ting Zhang, and Carmen C. Y. Poon∗

Abstract—Pulse transit time (PTT) is a cardiovascular parameter of emerging interest due to its potential to estimate blood pressure (BP) continuously and without a cuff. Both linear and nonlinear equations have been used in the estimation of BP based on PTT. This study, however, demonstrates that there is a hysteresis phenomenon between BP and PTT during and after dynamic exercise. A total of 46 subjects including 16 healthy subjects, 13 subjects with one or more cardiovascular risk factors, and 17 patients with cardiovascular disease underwent graded exercise stress test. PTT was measured from electrocardiogram and photoplethysmogram of the left index finger of the subject, i.e., a pathway that includes predominately aorta, brachial, and radial arteries. The results of this study showed that, for the same systolic BP (SBP), PTT measured during exercise was significantly larger than PTT measured during recovery for all subject groups. This hysteresis was further quantified as both normalized area bounded by the SBP–PTT relationship (AreaN) and SBP difference at PTT during peak exercise plus 20 ms (ΔSBP20). Significant attenuation of both AreaN (p < 0.05) and ΔSBP20 (p < 0.01) is observed in cardiovascular patients compared with healthy subjects, independent of resting BP. Since the SBP–PTT relationship are determined by the mechanical properties of arterial wall, which is predominately mediated by the sympathetic nervous system through altered vascular smooth muscle (VSM) tone during exercise, results of this study are consistent with the previous findings of autonomic nervous dysfunction in cardiovascular patients. We further conclude that VSM tone has a nonnegligible influence on the BP–PTT relationship and thus should be considered in the PTT-based BP estimation. Index Terms—Cuffless blood pressure (BP), mobile health, pulse transit time (PTT), vascular smooth muscle (VSM) tone, wearable device. Manuscript received March 28, 2013; accepted October 10, 2013. Date of publication October 23, 2013; date of current version January 16, 2014. This work was supported in part by the Hong Kong Innovation and Technology Commission (ITS/159/11) and in part by FP7 HeartCycle Project (FP7-216695), the National Basic Research Program 973 (no. 2010CB732606), and Guangdong Innovation Research Team Fund for LCHT in China. Asterisk indicates corresponding author. Q. Liu is with the Joint Research Centre for Biomedical Engineering, Department of Electronic Engineering, The Chinese University of Hong Kong, Shatin, Hong Kong (e-mail: [email protected]). Y.-T. Zhang is with the Joint Research Centre for Biomedical Engineering, Department of Electronic Engineering, The Chinese University of Hong Kong, Shatin, Hong Kongalso, and also with the Key Laboratory for Health Informatics, SIAT, Chinese Academy of Sciences, Shenzhen 518060, China (e-mail: [email protected]). B. P. Yan and C.-M. Yu are with the Division of Cardiology, Department of Medicine and Therapeutics, Prince of Wales Hospital, The Chinese University of Hong Kong, Shatin, Hong Kong (e-mail: [email protected]; [email protected]). ∗ C. C. Y. Poon is with the Department of Surgery and the Department of Electronic Engineering, The Chinese University of Hong Kong, Shatin, Hong Kong (e-mail: [email protected]). Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identifier 10.1109/TBME.2013.2286998

I. INTRODUCTION ARDIOVASCULAR diseases (CVDs) are the leading causes of deaths worldwide, bringing a great burden on individual, national and global economics [1]. The development of a variety of CVDs is associated with autonomic dysfunction. Specifically, declined vagal activity is a powerful indicator of overall mortality [2], while sympathetic overactivity is strongly related to diseases such as hypertension, obesity, and heart failure [3]–[5]. Pulse transit time (PTT) is a cardiovascular parameter of emerging interest due to its potential to estimate blood pressure (BP) continuously and without a cuff. Previous studies have shown that the PTT-based BP estimation has acceptable accuracy compared with gold-standard BP measurement methods in some situations [6], [7]. Nevertheless, a complex and nonlinear BP–PTT relationship has also been reported under some conditions [8], e.g., after the administration of vasoactive drugs [9] and during short-term physical exercise [10]. The BP–PTT relationship is dependent on the mechanical behavior of the arterial wall. Current PTT-based BP estimation methods are derived from the Moens–Korteweg equation under the assumption that artery is a passive, thin wall and purely elastic tube. The derived relationship between PTT and BP is approximately linear. Nevertheless, arterial wall in reality exhibits a specifically layered structure and comprises elastin, collagen, and smooth muscles. The three components function together to determine the overall mechanical behavior of artery. While elastin and collagen fibers response passively to the distending pressure, smooth muscle cells, which account for about 50% of wall compositions, can actively alter the mechanical property of the arterial wall by exerting an circumferential stress [11]. It has been reported that vasoactive drug induced vascular smooth muscle (VSM) contraction and dilation would change the elasticity of arteries [12]–[14]. Therefore, the VSM tone, which actually represents the activation level of VSM cells, is an essentially important factor that may influence the BP–PTT relationship. The VSM cell layer lies between autonomic nerve terminals and the endothelial cells; the VSM tone is accordingly under control of ANS and endothelial function. In fact, most blood vessels in human body do not have parasympathetic innervation, and thus are only influenced by the sympathetic nervous system [15]. There is a tonic balance between the release of vasodilating factors from the endothelium and vasoconstricting factors from sympathetic nerve terminals to maintain an appropriate VSM tone at rest [16]. Nevertheless, during exercise,

C

0018-9294 © 2013 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications standards/publications/rights/index.html for more information.

LIU et al.: ATTENUATION OF SYSTOLIC BLOOD PRESSURE AND PULSE TRANSIT TIME HYSTERESIS

347

alternations in ANS and endothelial function will occur, in order to adjust vascular impedance and reinforce blood flow redistribution [17]. Therefore, the influences of vasoactive drugs and physical exercise on BP–PTT relationship are both attributed to the VSM tone alternations under the circumstances. In this study, we presented hysteresis phenomenon between systolic BP (SBP) and PTT in human subjects who underwent an exercise protocol. Furthermore, we quantitatively evaluated the SBP–PTT hysteresis to examine whether the hysteresis is greater in healthy subjects than in cardiovascular patients as sympathetic overactivation have been previously reported in these patients. Fig. 1.

Typical recording of SBP and PTT during the experiment.

II. METHODOLOGY A. Study Population

C. Signal Processing

Sixty-four subjects were divided into three groups: 1) healthy subjects without known CVDs or risk factors (RFs) associated with CVDs, 2) subjects with one or more known RFs of CVDs, such as hypertension, diabetes mellitus, hyperlipidemia, and 3) patients with CVDs. The experiment was approved by the Joint Chinese University of Hong Kong—New Territories East Cluster Clinical Research Ethics Committee. All subjects were evaluated at the Division of Cardiology, Prince of Wales Hospital in Hong Kong. Inform consent was obtained from each subject before experiment.

Beat-to-beat SBP and PTT was extracted from Finapres BP, ECG, and PPG. SBP was defined as the maximal point of BP. PTT was measured as the time interval between the peaks of ECG R wave and the first derivative of PPG in the same cardiac cycle. SBP and PTT were then averaged every 60 beats for the hysteresis examination. Fig. 1 shows a typical recording of SBP and PTT time series during the experiment. Both the raw data and 60-beat averaged were presented.

B. Study Protocol The bicycle exercise stress tests were performed on the subjects in supine position. The experiment was conducted at least 1 h after meal in a standard patient room with temperature kept at 25 ◦ C. Upon arrival, the subject was asked to lie on a bed with his/her feet putting on a bicycle ergometer for 10 min to reach baseline. Cuff BP was measured by a registered nurse using auscultatory device at baseline. Continuous Finapres BP, electrocardiogram (ECG), cardiac output (CO), and photoplethysmogram (PPG) began to acquire until the end of the experiment. ECG and CO were measured by an impedance cardiographic device (Physio Flow PF-05, Macheren, France). PPG was obtained from the left index finger of the subject by using an in-house made acquisition device. After the 10-min rest, the bed was titled toward the left-hand side of the subject by 20◦ –30◦ and the subject was asked to start riding the bicycle at 25 W. The riding load was increased by 25 W every 2 min until it reached the tolerant limit of the subject. The load was then kept at the tolerant limit until the subject reached his/her target heart rate (HR) [85% × (220Age)]. For those subjects who cannot reach the target HR even at their exhausting condition, they are encouraged to keep riding to ensure their HR is as close to the target HR as possible. The subject stopped riding 1 min after reaching his/her maximal HR, then he/she was asked to lie still on the bed for recovery. The recovery phase lasted until CO dropped back to the baseline or at most for 15 min.

D. Hysteresis Examination The SBP–PTT hysteresis was characterized by higher SBP at a given PTT while BP is increasing during exercise and lower SBP at the same PTT while BP is decreasing during recovery. A qualified hysteresis should exhibit two clear boundaries during exercise and recovery, respectively. A representative example of qualified SBP–PTT hysteresis is shown in Fig. 2(a). Hysteresis was absent in 18 subjects where the SBP–PTT relationship was chaotic and the correlation was low due to subjects unable to keep on riding the bicycle as instructed. These 18 patients were excluded from further analysis. Finally, SBP–PTT data of 46 subjects were selected for the next hysteresis quantification. The evaluation approach was modified from the method proposed by another study for computation of QT–RR hysteresis [18]. Briefly, the method calculates the area between the two SBP–PTT regression curves during and after exercise, respectively. The quantification method is demonstrated in Fig. 2(b). For each subject, we first defined two segments: seg-Exe and seg-Rec on the SBP–PTT curve to represent the time period during and after exercise. As the relationship between SBP and PTT was not perfectly linear, the segments were determined manually according to the following criteria: 1) both seg-Exe and seg-Rec should comprise data points that increase or decrease as monotonically as possible; 2) data point in maximal exercise (either with highest SBP or shortest PTT) should be contained in the segments; and 3) SBP and PTT data in either segment should exhibit a moderate to high correlation. Both segments were then modeled by linear regression analysis. The hysteresis area noted as Area0 was computed as the

348


Fig. 2. (a) Representative example of qualified SBP–PTT hysteresis. (b) Demonstration of the quantification method of SBP–PTT hysteresis. Area0 is defined as the area between the exercise and recovery curves which were regressed from seg-Exe and seg-Rec, respectively. AreaN is the normalized area obtained from dividing Area0 by SBP range and PTT range. ΔSBP20 is the difference in SBP between exercise and recovery at PTT during peak exercise plus 20 ms.

Fig. 3. Processing method of the SBP–PTT hysteresis in subject 27. (a) Resampling the exercise/recovery curve to ten data points. (b) Normalizing the hysteresis to [0,1]. (c) Shifting the start point of exercise to (0, 0).

area between the two modeled curves over a range of PTT. As shown in Fig. 2(b), the range was from the smaller PTT of the minimal PTTs of the two segments; to the smaller PTT of the maximal PTTs of the two segments. The area was normalized by dividing SBP range and PTT range of the hysteresis to reach a normalized area—AreaN. Furthermore, since ANS alternations happens rapidly in late exercise and early recovery [18], the hysteresis was also quantified as the difference in SBP between exercise and recovery at a predetermined PTT, which is PTT during peak exercise plus 20 ms, toward better interpretation of the hysteresis. The parameter was defined as ΔSBP20. In addition, in order to calculate the group averaged hysteresis, the exercise/recovery curve was linearly resampled to ten data points from the start to the end of exercise/recovery for each subject, as shown in Fig. 3(a). Hysteresis was then averaged for each of the three groups. Toward better comparison of the hysteresis among groups, normalization was performed on each group by first normalizing the range of both PTT and BP to [0, 1] and then shifting the PTT–SBP curves to reinforce the

start point of exercise at (0, 0), as shown in Fig. 3(b) and (c), respectively. E. Statistics All data are presented as mean ± SD. Unpaired student’s t-test was employed to test the significance of differences in the hysteresis parameters between any two groups. III. RESULTS Table I summarizes the physical and hemodynamic characteristics of the selected subjects. There were 16 subjects in the healthy group (11 males, aged 59 ± 8 year), 13 subjects in the RF group (8 males, aged 56 ± 12 year), and 17 subjects in CVD group (15 males, aged 59 ± 10 year). The three groups have similar level of SBP and DBP at baseline. In addition, no significant difference was found among the groups in all the other characteristics. Table II gives the clinical diagnosis of subjects in RF group and CVD group. There were four, eight, and two patients in RF


TABLE I PHYSICAL AND HEMODYNAMIC CHARACTERISTICS OF SUBJECTS

TABLE II DIAGNOSIS OF SUBJECTS IN RF GROUP AND CVD GROUP

group and two, six, and four patients in CVD group have diabetes mellitus, hypertension, and hyperlipidemia, respectively. Furthermore, in CVD group, ten patients have coronary artery disease, seven patients have heart failure, one patient has peripheral arterial disease, and two patients have cerebrovascular disease. The group averaged hysteresis is shown in Fig. 4, with (a), (b), and (c) present absolute averaged hysteresis in healthy group, RF group, and CVD group, respectively. Fig. 5 gives the normalized hysteresis in (a) healthy group, (b) RF group, and (c) CVD group, respectively. The results demonstrate decreased hysteresis in RF group and CVD group compared with healthy group. Fig. 6 compares the two hysteresis quantifications: AreaN and ΔSBP20 among the three groups. AreaN was 0.32 ± 0.12 arbitrary unit (a.u.), 0.24 ± 0.14 a.u., and 0.22 ± 0.12 a.u.; ΔSBP20 was 26.58 ± 11.97 mmHg, 19.05 ± 15.40 mmHg, and 14.85 ± 10.12 mmHg for healthy, RF, and CVD groups, respectively. As revealed by the figure, both AreaN and ΔSBP20 demonstrate significant decrease in CVD group compared with healthy group (p < 0.05 for AreaN; p < 0.01 for ΔSBP20), indicating attenuated SBP–PTT hysteresis during exercise and recovery in cardiovascular patients.

349

IV. DISCUSSION In this experiment, subjects were asked to perform cycling exercise in a supine position while PPG was measured from their left index finger. The path which the pressure pulse travels through should be predominately aorta, brachial, and radial arteries. As leg cycling exercise was adopted in the protocol, vasodilation occurred mainly in the active limbs, to permit redistribution of blood flow to the contracting skeletal muscles. In contrast, since the left upper limb was kept still during the experiment, blood flow induced endothelial function changes in the PTT paths were assumed to be minimal; thus, the effects of endothelial function were regarded as negligible in this study. Therefore, although VSM tone is the result of interplay of ANS control and endothelial function, alternations of VSM tone during exercise and recovery in this study should be mainly attributed to the ANS control. It is hypothesized that if PTT was measured over the arteries of the exercising limbs, i.e., legs, the SBP–PTT hysteresis would be reversed with lower SBP during exercise and higher SBP during recovery at the same PTT, as vasodilation induced by the endothelial function plays a predominant role in these arteries during exercise. The hysteresis effect in BP and pulse wave velocity (PWV), which is the reciprocal of PTT, has been previously reported by in vivo dog experiment [19]. The study observed a shifted BP–PWV relationship in dogs when the neurohumoral regulatory factors were intact. However, the hysteresis phenomenon disappeared and BP–PWV relationship became “zigwag” when the nervous regulatory factor and part of the humoral regulatory factors were eliminated. It was, therefore, concluded that vasoconstriction and dilation were the main elements that cause hysteresis and further influence the relationship between BP and PWV. Specifically, for the same PWV, BP was higher during BP rises and lower during BP falls. In our study, SBP–PTT hysteresis is characterized by higher SBP at a given PTT while BP is increasing during exercise and lower SBP at the same PTT while BP is decreasing after exercise. Therefore, the direction of hysteresis characterized in our study is consistent with previous study. It is worth noting that the hysteresis effect between PTT and mean BP (MBP) as well as PTT and diastolic BP (DBP) were also observed in this study. By using the same quantification method, it can be obtained that the MBP–PTT and DBP– PTT hysteresis were also attenuated in cardiovascular patients, as shown in Fig. 7. Late exercise is associated with sympathoexcitation and minimal parasympathetic effects; while early recovery is characterized by rapid parasympathetic reactivation with persistent, but declining sympathoexcitation [18]. Since sympathetic and parasympathetic activities will lead to VSM contraction and dilation, respectively, it can be deduced intuitively that VSM tone is higher thus artery is stiffer during exercise than recovery. However, the hysteresis results in this study showed that, at a given SBP, PTT was longer (i.e., shorter PWV) during exercise than recovery. The paradox may be explained by taking into consideration of vessel radius decrease in consequence of vasoconstriction. It is evident that the increase in stiffness provided by smooth muscle activation is insufficient to offset the

350


Fig. 4. Averaged hysteresis in (a) healthy group, (b) RF group, and (c) CVD group. Solid circle (a), triangle (b), and square (c) represent exercise, and open circle (a), triangle (b), and square (c) indicate recovery; mean and standard deviation are both presented.

Fig. 5. Normalized hysteresis in (a) healthy group, (b) RF group, and (c) CVD group. Solid circle (a), triangle (b), and square (c) represent exercise, and open circle (a), triangle (b), and square (c) indicate recovery.

Fig. 6. Comparison of AreaN and ΔSBP20 among healthy group, RF group, and CVD group. ∗ : p < 0.05; ∗∗ : p < 0.01. (a.u.: arbitrary unit).

decrease in stiffness that results from contraction to a smaller radius [14]. Therefore, although VSM tone is enhanced during exercise, PWV compared as a function of pressure will decrease due to the reduced vessel radius. Furthermore, the results of this study showed that the SBP– PTT hysteresis during exercise and recovery was significantly attenuated in patients with CVDs. CVD development is a complicated process involving heighted sympathetic nervous system

activity and suppressed parasympathetic nervous system activity that impair the ability of the ANS to regulate the cardiovascular system [16]. Many studies have indicated augmented sympathetic nervous system activity and excessive vasoconstriction in heart failure patients [20], [21]. On the other hand, there is a close linkage between CVDs and heart failure. For example, most of the heart failure patients were found to have history of CVDs such as coronary heart disease and arterial hypertension [22]. Based on the aforementioned knowledge, the findings in this study could be attributed to the augmented sympathetic function and elevated basal VSM tone in such patients such that the SBP–PTT curve in recovery is closer to the curve during exercise when sympathoexcitation and vasoconstriction occurs. In addition, previous study has indicated a gradual increase in sympathetic activity in heart failure patients with normal BP, with either hypertension or obesity, and with both obesity and hypertension [23]; thus, the sympathetic tone seems to be gradually augmented during the pathological process of CVD. In this study, the hysteresis in subjects with one or more cardiovascular risk factors (RF group) was found to be reduced than that in healthy subjects (healthy group), while it was larger than that in CVD patients (CVD group). However, the differences were found to be not significant. The observation may be attributed


Fig. 7. Comparison of the hysteresis quantifications of the MBP–PTT and the DBP–PTT hysteresis among healthy group, RF group, and CVD group. ∗ : p < 0.05; ∗∗ : p < 0.01.

to the insufficient subject pool. On the other hand, the sympathetic nervous function may be partly preserved in subjects of RF group, leading to the insignificant results. In particular, Finpares was used for BP measurement in the study because it is currently one of the most widely acceptable methods that can measure BP noninvasively and continuously during dynamic exercise. Its specifically designed ‘‘Physiocal” algorithm is able to compensate and minimize the influences of vasoconstriction and vasodilation during exercise. More importantly, in this study, the SBP–PTT hysteresis was quantified by using the normalized hysteresis area (areaN) and SBP difference at the same PTT (ΔSBP20), it was accordingly the SBP changes, rather than the absolute SBP value, would affect the results. While its accuracy and precision was found to be sufficient when used to track BP changes [24], Finapres should be acceptable for measuring BP to obtain reliable results in the study. Nevertheless, PTT measured in this study is not the actual transit time of pressure pulse, but includes a pre-ejection period (PEP), which is a considerable time delay comprised of both the electromechanical delay and the isovolumic contraction period of left ventricle [9]. Although impedance cardiogram has been used to track PEP changes in this experiment, the signalto-noise ratio was found to be much lower than ECG, PPG, and BP during cycling exercise. As a result, if PEP was to be taken into account, the number of usable data segments in this study has to be substantially reduced. On the other hand, PEP is determined by the cardiac muscle contractility which is also controlled by the sympathetic nervous system. Resulted from heightened sympathetic tone, PEP will be shorten during ex-

351

ercise [25]. Therefore, if the PEP effect is removed, the actual SBP–PTT hysteresis should be enlarged, as an extra portion will be added to the hysteresis to account for the PEP shortening during exercise. In particular, previous study has demonstrated that the decrease in PEP during exercise was less in patients with cirrhosis than controls, due to the impaired sympathetic nervous function in the patients [25]. Based on these previous studies, it is concluded that the decrease in PEP during exercise is also less in cardiovascular patients than healthy subjects, as CVD development is usually related with sympathetic nervous system dysfunction. Consequently, the added portion accounting for the PEP effect should be larger, and the hysteresis phenomenon should be more significantly enhanced in healthy subjects than in CVD patients. From this perspective, though difficult to be measured during exercise, PEP will not affect the conclusion of this study. PTT is well known for its potential for cuffless BP estimation. However, its own clinical information has also been recognized for a long time. PTT is a potential screen tool for sleep related breathing disorders [26], [27]. It also offers a simple and noninvasive method to evaluate ANS effect on the cardiovascular system [28], [29]. Its reciprocal, PWV, is of great prognostic value as index of arterial stiffness and reported to predict a composite of cardiovascular outcomes above and beyond 24-h mean BP and other traditional CVD RFs, indicating that PTT contains information about the cardiovascular system in addition to BP. In this study, we demonstrate that although the baseline BP was similar in the three subject groups, hysteresis was significantly reduced in patients diagnosed with CVD. Although the recovery of HR immediately after exercise has been promoted as an index of parasympathetic function with important clinical implications [30], [31], convenient assessment of sympathetic activity which has a predominant role in cardiovascular control under a variety of situations, remains challenging [55]. The results of this study provide us a chance to examine solely sympathetic function by using indices quantifying the SBP–PTT hysteresis. This study revealed that VSM tone is a vitally important determinant of the BP–PTT relationship and therefore if PTT is to be used for BP estimation, the active VSM tone must be considered, especially scenarios when ANS varies due to drug effects, sleep, and exercise. Limitations: Data of 18 in 64 subjects are excluded in the analysis. In these subjects, SBP–PTT plots display chaotic relationship or overlapped loops mainly due to some subjects not able to continue cycling as instructed. In conclusion, this study presents a hysteresis effect between SBP and PTT during exercise and recovery. Quantitative analysis was applied on the SBP–PTT hysteresis in 46 subjects. The results showed that, in spite of similar level of resting BP, the SBP–PTT hysteresis was significantly attenuated in patients with CVD, suggesting an elevated sympathetic tone in these patients. We further concluded that VSM tone has a nonnegligible influence on the BP–PTT relationship, thus should be taken into account when using PTT as a surrogate for BP estimation.

352


ACKNOWLEDGMENT The authors would like to thank Dr. L. Wang and Ms. Y. B. Liu for helping with the collection of the data and Prof. G. Yip for giving advice on the design of the protocol of this study in the initial phase. REFERENCES [1] S. Mendis, P. Puska, and B. Norrving Eds., Global Atlas on Cardiovascular Disease Prevention and Control. Geneva, Switzerland: World Health Organization, 2011. [2] C. R. Cole, E. H. Blackstone, F. J. Pashkow, C. E. Snader, and M. S. Lauer, “Heart-rate recovery immediately after exercise as a predictor of mortality,” N. Engl. J. Med., vol. 341, no. 18, pp. 1351–1357, 1999. [3] J. N. Cohn, T. B. Levine, M. T. Olivari, V. Garberg, D. Lura, G. S. Francis, A. B. Simon, and T. Rector, “Plasma norepinephrine as a guide to prognosis in patients with chronic congestive heart failure,” N. Engl. J. Med., vol. 311, no. 13, pp. 819–823, 1984. [4] C. R. Benedict, B. Shelton, D. E. Johnstone, G. Francis, B. Greenberg, M. Konstam, J. L. Probstfield, and S. Yusuf, “Prognostic significance of plasma norepinephrine in patients with asymptomatic left ventricular dysfunction,” Circulation, vol. 94, no. 4, pp. 690–697, 1996. [5] S. C. Malpas, “Sympathetic nervous system overactivity and its role in the development of cardiovascular disease,” Physiol. Rev., vol. 90, no. 2, pp. 513–557, 2010. [6] C. C. Y. Poon and Y. T. Zhang, “Cuff-less and noninvasive measurements of arterial blood pressure by pulse transit time,” in Proc. IEEE 27th Annu. Int. Eng. Med. Biol. Soc., Shanghai, China, 2005, pp. 5877–5880. [7] M. Y. M. Wong, C. C. Y. Poon, and Y. T. Zhang, “An evaluation of the cuffless blood pressure estimation based on pulse transit time technique: A half year study on normotensive subjects,” Cardiovasc. Eng., vol. 9, no. 1, pp. 32–38, 2009. [8] G. Q. Zhang, M. W. Gao, D. Xu, N. B. Olivier, and R. Mukkamala, “Pulse arrival time is not an adequate surrogate for pulse transit time as a marker of blood pressure,” J. Appl. Physiol., vol. 111, no. 6, pp. 1681–1686, 2011. [9] R. A. Payne, C. N. Symeonides, D. J. Webb, and S. R. J. Maxwell, “Pulse transit time measured from the ECG: An unreliable marker of beat-to-beat blood pressure,” J. Appl. Physiol., vol. 100, no. 1, pp. 136–141, 2006. [10] J. Proenca, J. Muehlsteff, X. Aubert, and P. Carvalho, “Is pulse transit time a good indicator of blood pressure changes during short physical exercise in a young population?,” in Proc. IEEE 32nd Annu. Int. Conf. Eng. Med. Biol. Soc., Buenos Aires, Argentina, 2010, pp. 598–601. [11] W. W. Nichols and M. F. O’Rourke, McDonald’s Blood Flow in Arteries: Theoretic, Experimental, and Clinical Principles, 5th ed. London, U.K.: E. Arnold, 2005. [12] M. A. Zulliger, N. T. M. R. Kwak, T. Tsapikouni, and N. Stergiopulos, “Effects of longitudinal stretch on VSM tone and distensibility of muscular conduit arteries,” Amer. J. Physiol. Heart Circ. Physiol., vol. 283, no. 6, pp. H2599–H2605, 2002. [13] S. Kinlay, M. A. Creager, M. Fukumoto, H. Hikita, J. C. Fang, A. P. Selwyn, and P. Ganz, “Endothelium-derived nitric oxide regulates arterial elasticity in human arteries in vivo,” Hypertension, vol. 38, no. 5, pp. 1049–1053, 2001. [14] P. B. Dobrin and A. A. Rovick, “Influence of vascular smooth muscle on contractile mechanics and elasticity of arteries,” Amer. J. Physiol., vol. 217, no. 6, pp. 1644–1651, 1969. [15] R. Klabunde, Cardiovascular Physiology Concepts. Baltimore, MD, USA: Williams & Wilkins, 2011.

[16] K. F. Harris and K. A. Matthews, “Interactions between autonomic nervous system activity and endothelial function: A model for the development of cardiovascular disease,” Psychosom. Med., vol. 66, no. 2, pp. 153– 164, 2004. [17] M. H. Laughlin and D. H. Korzick, “Vascular smooth muscle: Integrator of vasoactive signals during exercise hyperemia,” Med. Sci. Sports Exerc., vol. 33, no. 1, pp. 81–91, Jan. 2001. [18] D. J. Pelchovitz, J. Ng, A. B. Chicos, D. W. Bergner, and J. J. Goldberger, “QT-RR hysteresis is caused by differential autonomic states during exercise and recovery,” Amer. J. Physiol. Heart Circ. Physiol., vol. 302, no. 12, pp. H2567–H2573, 2012. [19] W. Lu, H. Li, S. Tao, D. Zhang, Z. Jiang, L. Cui, J. Tu, and D. Gou, “Research on the main elements influencing blood pressure measurement by pulse wave velocity,” Front. Med. Biol. Eng., vol. 4, no. 3, pp. 189–199, 1992. [20] F. E. I. Cabrera, D. Bia, Y. Z´ocalo, and R. L. Armentano, “Smooth muscledependent changes in aortic wall dynamics during intra-aortic counterpulsation in an animal model of acute heart failure,” Int. J. Artif. Organs., vol. 32, no. 6, pp. 354–361, 2009. [21] G. S. Francis and J. N. Cohn, “The autonomic nervous system in congestive heart failure,” Annu. Rev. Med., vol. 37, no. 1, pp. 235–247, 1986. [22] G. Hasenfuss, “Animal models of human cardiovascular disease, heart failure and hypertrophy,” Cardiovasc. Res., vol. 39, no. 1, pp. 60–76, 1998. [23] G. Grassi, G. Seravalle, F. Quarti-Trevano, R. Dell’Oro, G. Bolla, and G. Mancia, “Effects of hypertension and obesity on the sympathetic activation of heart failure patients,” Hypertension, vol. 42, no. 5, pp. 873–877, 2003. [24] B. P. M. Imholz, W. Wieling, G. A. van Montfrans, and K. H. Wesseling, “Fifteen years experience with finger arterial pressure monitoring: Assessment of the technology,” Cardiovasc. Res., vol. 38, no. 3, pp. 605–616, Jun. 1998. [25] M. Bernardi, A. Rubboli, F. Trevisani, C. Cancellieri, A. Ligabue, M. Baraldini, and G. Gasbarrini, “Reduced cardiovascular responsiveness to exercise-induced sympathoadrenergic stimulation in patients with cirrhosis,” J. Hepatol., vol. 12, no. 2, pp. 207–216, 1991. [26] J. L. P´epin, N. Delavie, I. Pin, C. Deschaux, J. Argod, M. Bost, and P. Levy, “Pulse transit time improves detection of sleep respiratory events and microarousals in children,” Chest, vol. 127, no. 3, pp. 722–730, 2005. [27] S. E. Brietzke, E. S. Katz, and D. W. Roberson, “Pulse transit time as a screening test for pediatric sleep-related breathing disorders,” Arch. Otolaryngol. Head Neck Surg., vol. 133, no. 10, pp. 980–984, 2007. [28] T. Weiss, A. Del Bo, N. Reichek, and K. Engelman, “Pulse transit time in the analysis of autonomic nervous system effects on the cardiovascular system,” Psychophysiology, vol. 17, no. 2, pp. 202–207, 1980. [29] J. Y. A. Foo and C. S. Lim, “Pulse transit time as an indirect marker for variations in cardiovascular related reactivity,” Technol. Health Care, vol. 14, no. 2, pp. 97–108, 2006. [30] J. J. Goldberger, F. K. Le, M. Lahiri, P. J. Kannankeril, J. Ng, and A. H. Kadish, “Assessment of parasympathetic reactivation after exercise,” Amer. J. Physiol. Heart Circ. Physiol., vol. 290, no. 6, pp. H2446– H2452, 2006. [31] A. B. Chicos, P. J. Kannankeril, A. H. Kadish, and J. J. Goldberger, “Parasympathetic effects on cardiac electrophysiology during exercise and recovery in patients with left ventricular dysfunction,” Amer. J. Physiol. Heart Circ. Physiol., vol. 297, no. 2, pp. H743–H749, 2009.

Authors’ photographs and biographies not available at the time of publication.