Effect of Endurance Exercise Training and Curcumin Intake on Central Arterial Hemodynamics in Postmenopausal Women: Pilot Study

Arterial pressure in the central region (e.g., aorta and carotid artery) is composed mainly of the incident wave from the heart and the reflected wave from periphery. Age-related aortic stiffening and impaired endothelial function may result in the early return (in late systole) of the augmented reflection wave from periphery and thereby increase le# ventricular (LV) a#erload.1 As the increase in LV a#erload potentiates an increase in LV mass,2 an independent risk for heart failure and coronary heart disease mortality,3 central arterial hemodynamics has emerged as an important factor underlying the pathophysiology of cardiovascular disease.2,4–6 In this context, it should be noted that central arterial augmentation index (AIx), an index of LV a#erload, increases with advancing age although it changes less with aging in older individuals7,8 and is higher in women than similar aged men at all ages.8 In turn, older women may exhibit the highest central arterial AIx among men and women at all ages. $e importance of lifestyle modification, especially regular physical activity, to prevent the age-related increase in cardiovascular disease risks has been endorsed by major health organizations.9,10 In typical clinical settings, however, dietary modifications are usually prescribed along with regular exercise. In this context, our research group reported that a dietary intervention (e.g., daily lactotripeptides ingestion) with regular endurance exercise training has additive beneficial effect on vascular function (i.e., endothelial function, central arterial compliance) compared with either exercise training or dietary treatment alone in postmenopausal women.11,12 Recently, potential therapeutic or 1Human Technology Research Institute, National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba, Japan; 2Division of Sports Medicine, Graduate School of Comprehensive Human Sciences, University of Tsukuba, Tsukuba, Japan. Correspondence: Jun Sugawara (jun.sugawara@aist.go.jp)

Arterial pressure in the central region (e.g., aorta and carotid artery) is composed mainly of the incident wave from the heart and the reflected wave from periphery. Age-related aortic stiffening and impaired endothelial function may result in the early return (in late systole) of the augmented reflection wave from periphery and thereby increase le ventricular (LV) a erload. 1 As the increase in LV a erload potentiates an increase in LV mass, 2 an independent risk for heart failure and coronary heart disease mortality, 3 central arterial hemodynamics has emerged as an important factor underlying the pathophysiology of cardiovascular disease. 2,[4][5][6] In this context, it should be noted that central arterial augmentation index (AIx), an index of LV a erload, increases with advancing age although it changes less with aging in older individuals 7,8 and is higher in women than similar aged men at all ages. 8 In turn, older women may exhibit the highest central arterial AIx among men and women at all ages. e importance of lifestyle modification, especially regular physical activity, to prevent the age-related increase in cardiovascular disease risks has been endorsed by major health organizations. 9,10 In typical clinical settings, however, dietary modifications are usually prescribed along with regular exercise. In this context, our research group reported that a dietary intervention (e.g., daily lactotripeptides ingestion) with regular endurance exercise training has additive beneficial effect on vascular function (i.e., endothelial function, central arterial compliance) compared with either exercise training or dietary treatment alone in postmenopausal women. 11,12 Recently, potential therapeutic or ORIGINAL CONTRIBUTIONS Endurance Training, Curcumin, and Wave Reflection preventive effects of curcumin (diferuloylmethane), a yellow pigment in the spice turmeric, have been confirmed on various diseases, including pancreatic cancer, colon cancer, and Alzheimer's disease. 13,14 ese favorable effects seem to be associated with a wide range of beneficial properties of curcumin, including anti-inflammatory, antioxidant, chemopreventive, and chemotherapeutic activity presumably via multiple signaling pathways (i.e., survival pathways regulated by nuclear factor-κB, Akt, and growth factors; cytoprotective pathways; angiogenic pathways). 13,14 Antiinflammatory and antioxidant properties could potentially improve endothelial function and thereby attenuate the central wave reflection via the decrease in vascular tone. However, the effect on central arterial hemodynamic variables has never been addressed. Accordingly, the aim of this pilot study was to test the hypothesis that regular endurance exercise combined with daily curcumin ingestion improves central arterial hemodynamics (i.e., lowering LV a erload) to a greater extent than monotherapy with either intervention alone in postmenopausal women.

Subjects.
A randomized, double-blind, placebo-controlled, parallel design study was performed to test our hypothesis preliminarily. A total of 90 sedentary postmenopausal (e.g., more than 0.5 years since the last menses) women who were free of overt cardiovascular disease volunteered to participate.
irty-four candidates who were taking cardiovascular-acting medications including hormone replacement therapy and smokers were excluded. Random assignment with mean ageand brachial blood pressure (BP)-matching was done with the SORT function of Microso Excel. Fi y-six postmenopausal women were assigned randomly to one of the following interventions: "placebo ingestion" group (Pla), "curcumin ingestion" group (Cur), "exercise training with placebo ingestion" group (Ex+Pla), or "exercise training with curcumin ingestion" group (Ex+Cur). Final numbers of subjects who completed the intervention of exercise and/or curcumin intake and all measurements were 11 in Pla, 11 in Cur, 11 in Ex+Pla, and 12 in Ex+Cur, respectively (Figure 1). All subjects gave their written informed consent to participate. All procedures were reviewed and approved by the ethical committee of the University of Tsukuba.
Curcumin Ingestion. Curcumin or placebo was administered orally everyday for 8 weeks. Subjects in Cur and Ex+Cur groups ingested six pills of curcumin (per day), which consists 25 mg highly absorptive curcumin dispersed with colloidal nanoparticles 15 ( eracurumin; eravalues, Tokyo, Japan) (total 150 mg/day). Likewise, Pla and Ex+Pla groups ingested six placebo pills (per day) which were of similar shape and color and made of an equivalent dose of starch (e.g., dextrin and maltose). Subjects were instructed to record the number of ingested pills in a notebook everyday and not to alter their dietary habit (other than placebo or curcumin ingestion) throughout the intervention period.
Exercise Intervention. Subjects in Ex+Pla and Ex+Cur groups underwent endurance exercise training 3-6 days/ week for 8 weeks. e training program consists of supervised cycling exercise using electrically braked ergometer at the university 2-3 days/week and additional home-based walking exercise. Initially, subjects performed cycling or walking 25-30 min/ day, 3-4 days/week at a relatively low intensity of exercise, e.g., 60% of their individually determined peak heart rate (HR) which was evaluated at the incremental cycle exercise test (see below). As their exercise tolerance improved, the intensity and duration of endurance exercise were increased to 40-45 min/day, 4-5 days/week, at an intensity of 70-75% of peak HR. HR during exercise training was evaluated by a digital pulse rate monitor (SM-66; Skynie, Tokyo, Japan). Adherence to the exercise prescription was documented through the use of a uniaxial electrical accelerometer (Lifecorder; KENZ, Nagoya, Japan) and physical activity logs as described previously. 16 Subjects in Pla and Cur groups were instructed not to change their physical activity level.
Measurements. Blood biochemistry, arterial BP, HR, aortic pulse wave velocity (PWV), and peak oxygen consumption (VO 2 peak) were measured before and a er 8 weeks of each intervention. Subjects abstained from caffeine and fasted for at least 12 h before the experiment. Furthermore, subjects were studied 24 h a er their last exercise training session and/ or the last curcumin ingestion to avoid the immediate (acute) effects of either stimulus. A er 20-30 min of supine rest in a temperature-controlled dim room, cardiovascular function measurement followed blood sampling and aerobic capacity measurement were measured in each subject.  Recordings were made in triplicate with subjects in the supine position. Ankle-brachial BP index (ABI), a ratio between posterior-tibial and brachial arterial systolic BPs (SBPs), was evaluated to determine suspicious of peripheral arterial disease (ABI <0.9). None of subjects exhibited an ABI of <0.9, and were therefore allowed to participate in the protocol. Investigators who performed radial artery pressure waveforms recording with applanation tonometry and pressure waveform analysis were blinded to treatment (curcumin vs. placebo) and exercise. Radial arterial pressure waveforms were recorded by a validated applanation tonometry-based measurement device (Jentow; Colin Medical Technology) which was connected to an acquisition system (PowerLab 8/30; ADInstruments, Bella Vista, Australia) interfaced with a personal computer equipped with data acquisition so ware (LabChart 6; ADInstruments). Radial arterial pressure waveforms were calibrated with oscillometry-derived brachial BP and sampled at 1,000 Hz for off-line analysis. Stored radial arterial pressure waveforms were resampled at 128 Hz with data analysis so ware (AcqKnowledge; BIOPAC Systems, Santa Barbara, CA) 18 and then transferred into aortic pressure waveforms with an arterial waveform analysis so ware involving a validated generalized transfer function (SphygmoCor so ware; AtCor Medical, Sydney, Australia). 19 To quantify the magnitude of wave reflection from the periphery to the heart, augmented pressure (AP; peak pressure−pressure at the inflection point at systolic shoulder) was computed from synthesized aortic pressure waveforms. Aortic AIx was also calculated as AP divided by aortic pulse pressure. 20 AP and AIx were normalized for HR at 75 beats per minute (bpm) (AP 75 and AIx 75 , respectively). Aortic (carotid-femoral) PWV was calculated from the distance between two arterial recording sites divided by the transit time. Carotid and femoral arterial pulse waveforms were simultaneously recorded on the le common carotid and le common femoral arteries with a vascular testing device equipped with two applanation tonometry sensors (Form PWV/ABI; Colin Medical Technology). e transit time between carotid and femoral waveforms were automatically calculated with foot-to-foot method, as previously reported by our group. 21 Arterial path length was assumed the straight distance between two recording sites, which was obtained in triplicate with a random-zero measurement.

ORIGINAL CONTRIBUTIONS
Aerobic capacity. Statistical analyses. Treatment compliance of exercise training was compared by unpaired t-test between Ex+Pla and Ex+Cur. Treatment compliance of pill intake was evaluated by twoway factorial analysis of variance (ANOVA). To determine the effect of each intervention on all outcome measures, repeatedmeasures ANOVA and two-way factorial ANOVA were used. Analysis of covariance was used to determine the effect of each intervention on changes in aortic AP, AP 75 , AIx, AIx 75 a er eliminating influences of the expected confounding factors (e.g., changes in brachial BP, body mass, or blood cholesterols). In the case of a significant F value, a post hoc test using the Fisher's least significant difference test identified significant differences among mean values. All data are reported as means ± s.e.m. Statistical significance was set a priori at P < 0.05 for all comparisons.

RESULTS
ere were no significant differences in treatment compliance for placebo or curcumin intake among four groups (94.7 ± 3.7% in Pla; 99.0 ± 0.6% in Ex+Pla; 99.4 ± 0.3% in Cur; 98.7 ± 0.8% in Ex+Cur). No one reported any adverse side effects of curcumin ingestion. Average training frequencies of supervised cycling exercise were comparable between two groups (2.4 ± 0.2 in Ex+Pla vs. 2.4 ± 0.1 days/week in Ex+Cur, P = 0.93). In addition, there were no significant differences in the average training frequencies of either walking exercise (2.6 ± 0.3 in Ex+Pla vs. 2.1 ± 0.4 days/week in Ex+Cur, P = 0.66) or entire training (5.0 ± 0.3 in Ex+Pla vs. 4.3 ± 0.4 days/week in Ex+Cur, P = 0.19) between two groups. Table 1 shows selected physiological characteristics. ere were no significant group-differences in any baseline variables except for body mass: Ex+Cur group was significantly heavier than Pla group. A er the intervention, body mass and body mass index were decreased in Ex+Cur group (P < 0.05 for both). VO 2 peak increased a er the intervention in Ex+Pla (P < 0.05) but not in the other groups. A er the intervention, plasma levels of high-density lipoprotein cholesterol also increased in Ex+Pla (P < 0.05) but not in the other groups. Plasma level of low-density lipoprotein cholesterol did not change with the intervention in all groups. HR and ejection duration did not change with the intervention in all groups. A er the intervention, brachial SBP decreased in Ex+Pla and Ex+Cur (P < 0.05 for both). In addition, Ex+Cur group demonstrated significant reductions in brachial mean arterial pressure and diastolic BP (P < 0.05 for both). Brachial pulse pressure did not change with the intervention in all groups. Table 2 summarizes the response of aortic BP, central (e.g., aortic and carotid arterial) and peripheral (e.g., radial arterial) AIx, and aortic PWV to the intervention. ere were no significant ORIGINAL CONTRIBUTIONS Endurance Training, Curcumin, and Wave Reflection group-differences in baseline variables. Aortic SBP and diastolic BP significantly decreased in Ex+Cur group but not in the other groups. Aortic pulse pressure did not change with the intervention in all groups. ere were no significant changes in aortic AP and AIx throughout the intervention, whereas the HR-corrected values (e.g., aortic AP 75 and AIx 75 ) were significantly decreased in Ex+Cur group but not in the other groups. Carotid AIx did not change significantly in each group. Radial AIx decreased in Ex+Cur group, albeit not significant. Consequently, radial AIx was significantly lower in Ex+Cur group compared with Pla and Ex+Pla groups a er the intervention.
As shown in Figure 2, the magnitudes of reduction in AP 75 and AIx 75 were significantly greater in Ex+Cur group compared with those in Pla and Ex+Pla groups, respectively. ese differences remained significant when the influences of the expected confounding factors (i.e., changes in brachial mean arterial pressure, body mass, or blood cholesterols) were statistically excluded. Aortic PWV did not change significantly with the intervention in all groups.

DISCUSSION
e salient findings of this pilot study are: (i) aortic SBP and HR-corrected aortic AP and AIx significantly decreased with the combined treatment of regular endurance exercise intervention and daily curcumin ingestion but did not change sig-nificantly with either regular endurance training or curcumin ingestion alone and (ii) the reduction in HR-corrected aortic AP and AIx with the combination of regular exercise training and curcumin ingestion was significantly larger than those of other groups even when the influences of confounding factors (i.e., mean arterial pressure, body weight, metabolic risks) were statistically excluded. Our findings may provide initial preliminary evidence that regular endurance exercise combined with daily curcumin ingestion may reduce LV a erload to a greater extent than monotherapy with either intervention alone in postmenopausal women.
In the present study, aortic and carotid AIx did not change across the intervention period, whereas aortic AIx corrected by HR decreased significantly with the combined treatment of regular endurance exercise training and daily curcumin ingestion. Although the concomitant change in HR was small (−2 bpm), it might mask the lowering effect on aortic AIx (non-adjusted value). On the other hand, in Ex+Pla group, aortic AIx did not change significantly even when it was corrected by HR. Tanaka et al. 22 indicated in the cross-sectional study that endurance-trained (average 13 years of training experience) postmenopausal women have significantly lower carotid AIx compared with sedentary peers. A cause of this inconsistency is unclear but likely to be associated with the short duration of training intervention. Alternatively, it might Data are mean ± s.e.m. Cur, curcumin; DBP, diastolic blood pressure; Ex+Cur, exercise training+curcumin; Ex+Pla, exercise training+placebo; HDL, high-density lipoprotein; LDL, low-density lipoprotein; MAP, mean arterial pressure; Pla, placebo; PP, pulse pressure; SBP, systolic blood pressure; VO 2 peak, peak oxygen uptake. *P < 0.05 vs. before intervention; **P < 0.05 vs. Pla group.

ORIGINAL CONTRIBUTIONS
Endurance Training, Curcumin, and Wave Reflection be due to the small sample size. However, our findings suggest that the daily curcumin intake seems to have additive impact in lowering BP by regular endurance training in normotensive postmenopausal women. We identified the attenuated magnitude of reflected wave (e.g., AP 75 ) in Ex+Cur group, whereas aortic PWV and ejection duration, expected determinants of aortic AIx, did not change significantly throughout the intervention period.
ese results may suggest that the decreased aortic AIx may be attributed to the attenuation of the central wave reflection, although recently the notion that reflected waves are important contributors to augmenting central BP has been debated. [23][24][25][26] To determine the exact underlying mechanism(s), contribution of other possible factors (i.e., arterial reservoir pressure) has to be considered. e results of the present pilot study suggest that in normotensive postmenopausal women the combined endurance exercise training and curcumin treatment may have more favorable effects on central arterial hemodynamic (i.e., lowering LV a erload) compared with either endurance exercise training or curcumin treatment alone.

Study limitation
ere are several limitations to our current protocol. First, aortic BP was estimated from radial arterial waveform by applying general transfer function. e use of transfer functions to estimate central pressure from peripheral pressure has been a topic

ORIGINAL CONTRIBUTIONS
Endurance Training, Curcumin, and Wave Reflection of debate (i.e., existence of individual difference). 27,28 Second, this pilot study has a small sample size because we rigorously screened candidates to determine the effect of primary prevention on age-related increase in BP in postmenopausal women. ereby, statistical power provided by a post hoc power analysis was not always high enough (i.e., α >0.80). Future studies should be warranted in a larger sample size and different populations, such as elderly men and patients with hypertension. ird, we applied the combination of supervised cycling exercise and additional home-based walking exercise in order to maintain exercise training compliance of subjects, which is based on our experience of training intervention studies. erefore, we might not control for cycling and walking training frequencies on each participant. Potential influence of divergent training compliance on trainability of aerobic capacity or changes in central arterial hemodynamics could not be completely ruled out.