Cardiovascular function in healthy Himalayan high-altitude dwellers

Atherosclerosis 236 (2014) 47e53

Contents lists available at ScienceDirect

Atherosclerosis journal homepage: www.elsevier.com/locate/atherosclerosis

Cardiovascular function in healthy Himalayan high-altitude dwellers R.M. Bruno a, b, *, A. Cogo c, L. Ghiadoni b, E. Duo c, L. Pomidori c, R. Sharma e, G.B. Thapa e, B. Basnyat e, M. Bartesaghi d, E. Picano a, R. Sicari a, S. Taddei b, L. Pratali a a

Institute of Clinical Physiology e CNR, Pisa, Italy Department of Clinical and Experimental Medicine, University of Pisa, Italy c Biomedical Sport Studies Center, University of Ferrara, Italy d Department of Experimental Medicine, Laboratory of Clinical Physiology and Sport Medicine, University of Milano-Bicocca, Italy e Nepal International Clinic, Kathmandu, Nepal b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 15 January 2014 Received in revised form 11 June 2014 Accepted 19 June 2014 Available online 25 June 2014

Background: Residents of the Himalayan valleys uniquely adapted to their hypoxic environment in terms of pulmonary vasculature, but their systemic vascular function is still largely unexplored. The aim of the study was to investigate vascular function and structure in rural Sherpa population, permanently living at high altitude in Nepal (HA), in comparison with control Caucasian subjects (C) living at sea level. Methods and results: 95 HA and 64 C were enrolled. Cardiac ultrasound, flow-mediated dilation (FMD) of the brachial artery, carotid geometry and stiffness, and aortic pulse wave velocity (PWV) were performed. The same protocol was repeated in 11 HA with reduced FMD, after 1-h 100% O2 administration. HA presented lower FMD (5.18 ± 3.10 vs. 6.44 ± 2.91%, p ¼ 0.02) and hyperemic velocity than C (0.61 ± 0.24 vs. 0.75 ± 0.28 m/s, p ¼ 0.008), while systolic pulmonary pressure was higher (29.4 ± 5.5 vs. 23.6 ± 4.8 mmHg, p < 0.0001). In multiple regression analysis performed in HA, hyperemic velocity remained an independent predictor of FMD, after adjustment for baseline brachial artery diameter, room temperature and pulse pressure, explaining 8.7% of its variance. On the contrary, in C brachial artery diameter remained the only independent predictor of FMD, after adjustment for confounders. HA presented also lower carotid IMT than C (0.509 ± 0.121 vs. 0.576 ± 0.122 mm, p < 0.0001), higher diameter (6.98 ± 1.07 vs. 6.81 ± 0.85 mm, p ¼ 0.004 adjusted for body surface area) and circumferential wall stress (67.6 ± 13.1 vs. 56.4 ± 16.0 kPa, p < 0.0001), while PWV was similar. O2 administration did not modify vascular variables. Conclusions: HA exhibit reduced NO-mediated dilation in the brachial artery, which is associated to reduced hyperemic response, indicating microcirculatory dysfunction. A peculiar carotid phenotype, characterized by reduced IMT and enlarged diameter, was also found. © 2014 Elsevier Ireland Ltd. All rights reserved.

Keywords: High altitude Endothelial function Arterial stiffness Carotid remodeling Echocardiography Hypoxia

1. Introduction Many residents of the Himalayan valleys and Tibetan Plateau live at high altitude, experiencing O2 concentrations that are about 40% lower than those at sea level. Compared to other populations living at high altitude, such as Andean populations, they developed a favorable phenotype, characterized by lower prevalence of pulmonary hypertension and polycythemia despite decreased arterial O2 content [1e5]. In particular, O2 delivery to the cells is suspected to be maintained by compensative modulation of pulmonary vascular flow, probably due to tonically elevated pulmonary NO * Corresponding author. Institute of Clinical Physiology e CNR, Via Moruzzi 1, 56124 Pisa, Italy. Tel.: þ39 (0) 50 3152377; fax: þ39 (0) 50 3152355. E-mail address: [email protected] (R.M. Bruno). http://dx.doi.org/10.1016/j.atherosclerosis.2014.06.017 0021-9150/© 2014 Elsevier Ireland Ltd. All rights reserved.

production [6,7]. NO is a key molecule in systemic and pulmonary vascular physiology, for its vasodilating, antithrombotic and antimitotic properties [8]. Reduced NO availability in the systemic circulation, which is the main feature of endothelial dysfunction, has been recognized as the first step towards development of atherosclerosis [8]. Acute hypoxia can induce endothelial dysfunction and activation in individuals living at sea-level [9,10]. Furthermore, diseases characterized by chronic hypoxia show reduced NOmediated vasodilation [11]. However, systemic vascular characteristics of populations chronically exposed to hypobaric hypoxia are still unknown. We hypothesized that chronic exposure to hypobaric hypoxia might impair systemic endothelial function, thus favoring the development of cardiovascular damage. Accordingly, the aim of the study was to investigate the presence of reduced NO-mediated vasodilation in rural Sherpa population,

48

R.M. Bruno et al. / Atherosclerosis 236 (2014) 47e53

permanently living in Khumbu Valley (Nepal) at high altitude. In order to assess the presence of subclinical cardiovascular damage and its association with endothelial dysfunction, we measured a panel of established biomarkers (aortic and carotid stiffness, carotid intima-media thickness, left ventricular mass and diastolic function). Furthermore, pulmonary pressures were estimated, to evaluate response to hypoxia in the pulmonary vasculature. 2. Methods 2.1. Study population The study was part of the SHARE project (Stations at High Altitude for Research on the Environment). The study population was constituted by 95 high-altitude dwellers (HA), born and permanently living in the Khumbu Valley (Nepal), enrolled by local advertising in three rural villages (altitude 2600, 3800 and 3800 m respectively). Criteria of inclusion were age between 15 and 65 years, apparent good health status, and written informed consent. Criteria of exclusion were known established cardiovascular or renal disease, cardiovascular risk factors or treatments, active infections or neoplasm, pregnancy. We compared the vascular features of HA with those of 64 Caucasian subjects (C), living and studied at the sea-level in Italy, recruited according to the same criteria, and matched for age, sex, mean blood pressure (BP), and body mass index (BMI). All the subjects enrolled were aware of the purposes of the study and gave written informed consent. The study was conducted with the approval of the Ethical Committee in Italy and of the Nepal Health Research Council (NHRC) (Kathmandu, Nepal) and of the Nepal Academy of Science and Technology (NAST) (Clinical Trials Gov Registration #NCT01329159). An extended version of the Methods section is available as Online-only Supplement. 2.2. Experimental protocol All measurements were performed in the morning after an overnight fasting, in a quiet room. Medical history was collected by Nepalese-speaking physicians (R.S. and G.B.T.). Brachial BP was measured with the individual resting in a supine position for at least 10 min, three times at 2-min intervals, and averaged on the last two measurements. Finger O2 saturation (SO2), weight and height were also taken.

by sublingual administration of 25 mg glyceryl-trinitrate (GTN). Flow velocity (FV) was recorded throughout the recording, and baseline and hyperemic FV were computed. Carotid geometry and stiffness variables were assessed by the automated analysis of common carotid ultrasound scans [19]. Carotid distension (DD), cross-sectional distensibility coefficient (DC) and compliance coefficient (CC) were calculated. Common carotid intima-media thickness (IMT) was automatically measured on the same image sequences, and wall to lumen ratio (W/L) and static circumferential wall stress were computed [20]. Carotid-femoral and carotid-radial pulse wave velocity (PWV) were assessed by applanation tonometry (PulsePen, Diatecne: Milan, Italy), according to international recommendations, as previously described [17]. Carotid systolic BP and pulse pressure (PP) were then obtained from carotid pressure waveform, using brachial BP for calibration, as well as carotid augmented pressure and augmentation index [21]. 2.5. Effect of O2 administration In order to assess the role of hypoxia per se in inducing endothelial dysfunction, in 11 subjects with reduced FMD (below the median value in the HA population), the protocol was repeated after 100% O2 administration for 1 h, titrated to maintain SO2 around 100%. 2.6. Statistical analysis Statistical analysis was performed using NCSS 2008 (NCSS: Kaysville, Utah, USA). Results were expressed as mean ± SD. Differences in means among groups were analyzed using ANOVA for normally distributed variables, or KruskaleWallis Z Test for not normally distributed variables. Analysis of covariance was also used to compare vascular parameters, when indicated. An ANCOVAbased allometric approach was used in order to adjust for the influence of baseline diameter on FMD [22]. Categorical variables were analyzed by c2 test. Spearman's rank was used to explore correlations among variables. Multiple linear regression analysis was performed including parameters correlated with the dependent variable (FMD) with p < 0.10. 3. Results

2.3. Echocardiography Left ventricular (LV) dimensions were taken and used to calculated LV mass [12]. LV ejection fraction (EF) was calculated by the modified biplane Simpson's method [12], while cardiac output was calculated from LV outflow tract diameter and time-velocity integral [13]. Doppler mitral E flow-velocity wave and tissue Doppler mitral annulus flow e0 early diastolic velocity were acquired for the calculation of E/e0 . Systolic pulmonary artery pressure (PAP) was estimated from a trans-tricuspid gradient (right atrium e right ventricle gradient, RA-RV gradient) calculated from the maximal velocity of continuous Doppler tricuspid regurgitation [13,14]. Mean PAP, left atrial pressure (LAP) [15] and pulmonary vascular resistance (PVR) were also calculated [16]. 2.4. Vascular function and structure Endothelium-dependent response was assessed by ultrasound as increase of the brachial artery diameter (BAD) in response to increased blood flow (flow-mediated dilation, FMD), as previously described [17,18]. Endothelium-independent dilation was obtained

3.1. Clinical and echocardiographic characteristics of the study population As expected, HA had higher heart rate and lower SO2 and body surface area (BSA) than C, but similar BMI. They also showed higher diastolic and lower systolic BP values, leading to a lower PP, in the presence of similar mean BP values, HA. Room temperature during the experimental sessions was significantly lower in HA than in C (Table 1). LV systolic diameters and wall thickness corrected for body surface area, were not significantly different in HA and C, resulting in a similar LV mass index. LV diastolic diameter was significantly reduced in HA, but significance was lost upon adjustment for BSA and heart rate (p ¼ 0.10). EF and cardiac output and E/e0 were comparable in the two groups under investigation. Systolic and mean PAP were significantly higher in HA than in C (Table 1), with 15 (15.8%) and 7 (7.4%) individuals presenting pulmonary hypertension, defined with a cut-off of 35 and 25 mmHg respectively. Also PVR were increased in HA as compared with C, with 19 (20.0%) individuals with PVR3 mmHg/min/L.

R.M. Bruno et al. / Atherosclerosis 236 (2014) 47e53 Table 1 Clinical and echocardiographic characteristics of the study population.

Men (n, %) Age (years, range) SO2 (%) Room temperature ( C) Weight (kg) Height (m) Body surface area (m2) BMI (kg/m) Systolic BP (mmHg) Diastolic BP (mmHg) Mean BP (mmHg) PP (mmHg) Heart rate (bpm) LV end-diastolic diameter (mm) LV end-systolic diameter (mm) Septum thickness (mm) Posterior wall thickness (mm) LV mass index (g/m2) Cardiac output (L/min) EF (%) E/e’ mean LAP (mmHg) RV/RA gradient (mmHg) Systolic PAP (mmHg) Mean PAP (mmHg) PVR (mmHg/min/L) a

49

Table 2 Vascular characteristics of the study population.

Caucasian controls e C (n ¼ 64)

Himalayan dwellers e HA (n ¼ 95)

p value

24, 37.5% 36.2 ± 12.4 (17e63) 98.2 ± 0.91 23 ± 1

30, 31.6% 33.7 ± 13.8 (15e65) 90.6 ± 2.6 18 ± 2

0.44 0.22

65.0 ± 12.3 1.69 ± 0.09 1.74 ± 0.33

57.5 ± 8.3 1.57 ± 0.08 1.58 ± 0.14

<0.0001 <0.0001 <0.0001

22.7 ± 3.0 119.7 ± 11.5 69.9 ± 8.1 86.5 ± 8.1 49.9 ± 9.9 66.7 ± 12.2 45 ± 5

23.3 ± 2.9 113.2 ± 11.6 76.2 ± 8.9 88.5 ± 8.8 36.8 ± 9.8 73.3 ± 12.7 41 ± 4

0.19 0.0005 <0.0001 0.13 <0.0001 0.001 0.03a

26 ± 5

25 ± 5

0.68a

9.2 ± 1.5

8.3 ± 1.3

0.71a

8.4 ± 1.3

8.0 ± 1.1

0.83a

72 ± 17 4.6 ± 1.1

65 ± 16 4.9 ± 1.7

0.07 0.39

63.1 ± 6.7 5.71 ± 1.6 8.3 ± 1.7 18.6 ± 4.8 23.6 ± 4.8 16.1 ± 2.9 1.86 ± 0.82

65.3 ± 6.6 6.39 ± 2.1 9.3 ± 2.6 24.5 ± 5.4 29.4 ± 5.5 19.7 ± 3.3 2.30 ± 1.21

0,14 0.09 0.02 <0.0001 <0.0001 <0.0001 0.003

<0.0001 0.02

p value obtained by ANCOVA, considering body surface area as covariate.

3.2. Endothelial function in the brachial artery BAD and FMD acquisition were successful in 57 C and 88 HA subjects, whereas FV measurement was available in 35 C and in 55 HA subjects for technical problems during the first expedition; clinical characteristics of this subgroup were largely superimposable to those of the overall population, and are shown in Table 1 of the Online supplement. Baseline BAD and FV were similar in the two groups; when considering body surface area as covariate, baseline BAD tended to be greater in HA than in C (p ¼ 0.08). HA presented reduced FMD and hyperemic FV than C (Table 2). A reduced endothelial function in HA was confirmed also by allometric analysis. Log-transformed difference between baseline and peak BAD (lnBAD-difference), considering log-transformed baseline BAD (ln-BAD) as covariate, was significantly lower in HA as compared to C (corrected FMD 4.91 ± 0.51 vs. 6.66 ± 0.67%, p ¼ 0.026). In the subgroup of individuals in whom hyperemic FV was available, FMD tended to be lower in HA than in C even considering hyperemic FV as a covariate (5.03% vs. 6.02%, p ¼ 0.08). Brachial artery response to GTN was significantly increased in HA (Table 2). In C univariate analysis showed that FMD was significantly correlated with body surface area (r ¼ 0.306, p ¼ 0.046), PP (r ¼ 0.379, p ¼ 0.008), baseline BAD (r ¼ 0.693, p < 0.001, Fig. 1a) and tended to correlate with age (r ¼ 0.240, p ¼ 0.099). On the other hand, no significant correlation was found between FMD and hyperemic FV (r ¼ 0.021, p ¼ 0.906, Fig. 1b). In HA univariate analysis showed that FMD was significantly correlated with PP (r ¼ 0.225, p ¼ 0.036), baseline BAD

Caucasian controls e C (n ¼ 64) Brachial artery diameter (mm)a FMD (%)a Baseline FV (m/s)b Hyperemic FV (m/s)b Response to GTN (%)a Carotid-femoral PWV (m/s) Carotid-radial PWV (m/s) Carotid IMT (mm) Mean carotid diameter (mm) Distension (mm) Wall to lumen ratio Circumferential wall stress (kPa) Carotid compliance (m2*kPa1) Carotid distensibility (kPa1) Carotid PP (mmHg) Augmented Pressure (mmHg) Augmentation Index (%) Young's elastic modulus (kPa) a b

3.57 6.44 0.16 0.75 6.90 6.90 7.86 0.576 6.81 0.57 0.17 56.4 1.07 33.2 44.3 5.7 6.7 0.31

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.77 2.91 0.08 0.28 2.47 1.73 2.28 0.122 0.85 0.13 0.04 16.0 0.68 12.3 10.5 3.2 8.8 0.09

Himalayan dwellers e HA (n ¼ 95)

p value

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.47 0.02 0.70 0.008 0.04 0.83 0.09 <0.0001 0.48 0.06 <0.0001 <0.0001 0.04 0.07 <0.0001 0.92 0.21 0.62

3.64 5.18 0.17 0.61 8.21 7.04 8.76 0.509 6.98 0.53 0.15 67.6 1.24 38.2 36.7 6.5 10.8 0.35

0.74 3.10 0.08 0.24 3.11 2.24 2.08 0.121 1.07 0.10 0.02 13.1 0.63 17.9 12.1 5.4 14.2 0.10

Data available for 57C and 88 HA. Data available for 35C and 55 HA.

(r ¼ 0.216, p ¼ 0.043, Fig. 1a), hyperemic FV (r ¼ 0.343, p ¼ 0.011, Fig. 1b) and tended to correlate with room temperature (r ¼ 0.206, p ¼ 0.054). Multiple regression models were built in order to investigate independent determinants of FMD in the two populations (Table 3). In Model 1, adjusted for vascular determinants of FMD (baseline BAD and FV-difference), BAD remained an independent predictor of FMD in C (Table 3). This was confirmed also by Model 2, including the other confounders. On the contrary in HA, multiple regression confirmed that FV-difference was significantly correlated with FMD, regardless of baseline BAD in Model 1 (Table 3). In Model 2, FV-difference and room temperature remained independent predictors of FMD, after adjustment for baseline BAD and PP (Table 3). Superimposable results were obtained both in HA and SL in the allometric models, when lnBAD-difference was considered as dependent variable, with lnBAD among confounding factors (Table 3). We also explored the relationship between endothelial function and variables connected to adaptation to hypoxia or with established markers of cardiac and vascular damage. Among HA, FMD was not correlated to systolic and mean PAP (r ¼ 0.150, p ¼ 0.162 for both) and PVR (r ¼ 0.159, p ¼ 0.141), as well as to SO2 (r ¼ 0.034, p ¼ 0.752) and heart rate (r ¼ 0.051, p ¼ 0.632). Similar results were obtained in the C group (data not shown). Conversely, in HA FMD correlated with LV mass index (r ¼ 0.288, p ¼ 0.009) and tended to correlate with IMT (r ¼ 0.177, p ¼ 0.099), with superimposable results in the C group (LV mass index: r ¼ 0.431, p ¼ 0.036 and IMT r ¼ 0.276, p ¼ 0.063). 3.3. Arterial geometry and stiffness HA showed a significantly greater carotid diameter in comparison to C when BSA was considered as covariate (p ¼ 0.004). Carotid IMT was significantly reduced in HA subjects, leading to a reduced wall to lumen ratio (W/L) and an increased static circumferential wall stress (Fig. 2). Both carotid PP and distension were lower in HA than in C. Carotid compliance was significantly higher and carotid distensibility tended to be greater in HA; conversely, there were no significant differences between HA and C as far as carotid stiffness, carotid-femoral and carotid-radial PWV were concerned (Table 2). Carotid-femoral PWV was not significantly different between the

50

R.M. Bruno et al. / Atherosclerosis 236 (2014) 47e53

Fig. 1. Scatter plots representing the relationship of flow-mediated dilation (FMD) with brachial artery diameter (a) and hyperemic flow velocity (FV) (b) in Caucasian sea-level controls (gray circles) and Himalayan high-altitude individuals (black circles).

two groups even after inserting mean BP and age as a covariate (p ¼ 0.23); similar results were obtained for carotid stiffness parameters (compliance: p ¼ 0.56; distensibility: p ¼ 0.14; stiffness p ¼ 0.75). Augmented pressure and Augmentation index were similar in HA and C even after considering age, mean BP, height and heart rate as covariates (p ¼ 0.28 and p ¼ 0.80 respectively). 3.4. Effect of O2 administration In the 11 HA studied, SO2 rose from 89 ± 2 to 99 ± 1% after O2 administration (p < 0.0001). This was accompanied by a significant reduction in heart rate (from 72 ± 14 to 55 ± 9 bpm, p ¼ 0.0003) and by unchanged BP (from 114 ± 13/75 ± 9 to 109 ± 18/71 ± 9 mmHg, p ¼ ns). Corrected FMD, obtained by allometric scaling analysis, was not modified by O2 administration (from 2.6 ± 0.7 to 2.5 ± 0.7%, p ¼ 0.84), as well as baseline BAD and hyperemic FV. An increase in augmentation index was observed (from 7.9 ± 8.2 to 16.3 ± 11.1%, Table 3 Multiple regression models exploring determinants of endothelial function in C and HA. Variable Caucasian sea-level controls (C) Dependent Model 1 variable FMD (0.524 full r2) Model 2 (0.579 full r2)

Baseline BAD Hyperemic FV Baseline BAD Hyperemic FV age PP BSA Dependent variable Model 1 lnBAD-bas lnBAD-diff (0.538 full r2) Hyperemic FV Model 2 lnBAD-bas (0.608 full r2) Hyperemic FV age PP BSA Himalayan high-altitude individuals (HA) Dependent Model 1 Baseline BAD variable FMD (0.134 full r2) Hyperemic FV Model 2 Baseline BAD (0.262 full r2) Hyperemic FV room temperature PP Model 1 Dependent lnBAD-bas (0.140 full r2) variable Hyperemic FV lnBAD-diff Model 2 lnBAD-bas (0.265 full r2) Hyperemic FV room temperature PP

r2

p value

0.516 0.008 0.472 0.000 0.014 0.057 0.036 0.537 0.001 0.500 0.005 0.011 0.059 0.033

<0.0001 0.473 <0.0001 0.690 0.661 0.673 0.210 <0.0001 0.361 <0.0001 0.798 0.757 0.803 0.149

0.060 0.074 0.048 0.095 0.111 0.008 0.048 0.092 0.036 0.115 0.107 0.007

0.107 0.041 0.165 0.016 0.013 0.360 0.098 0.037 0.130 0.015 0.014 0.389

p ¼ 0.005), but was not significant when heart rate was considered as covariate (p ¼ 0.92). The remaining vascular parameters were not modified by O2 administration.

4. Discussion This study demonstrated the presence of a unique cardiovascular phenotype in Himalayan healthy subjects born and permanently living at high altitude, free of traditional cardiovascular risk factors. NO plays a key role in mediating acute hypoxic vasodilation in healthy humans in resistance vessels [23,24], and seems to be involved also in chronic adaptation to hypoxia. High-altitude natives, such as Tibetans and Nepalese Sherpa, developed genetic adaptations to their hypoxic environment, as recently demonstrated [25,26]. In particular, lower vascular resistances [27], which have been attributed to tonically elevated NO production [1,6,7], have been demonstrated in the peripheral microcirculation. The present study demonstrated that a chronic vasodilation is present also in muscular and musculo-elastic arteries, as indicated by the presence of an increased BSA-corrected brachial and carotid diameter. Thus our results support the hypothesis that a chronically vasodilated microcirculation and macrocirculation may allow Sherpas to guarantee a sufficient tissue oxygenation in spite of a reduced arterial O2 content. On the other hand, to our knowledge stimulated NO release has never been explored in this population. To date, the effect of hypoxia on the human endothelium is matter of debate, with reduced function reported in individuals prone to high-altitude pulmonary edema (HAPE) after acute normobaric hypoxia [9], in patients with obstructive sleep apnea syndrome [11], and in Aymara highaltitude dwellers with pulmonary hypertension [28]. Conversely, a preserved function was reported in Aymara healthy high-altitude dwellers [28] and in HAPE-resistant individuals after acute hypoxia [9]. Our study demonstrated for the first time a reduced FMD in the brachial arteries of Himalayan healthy subjects born and permanently living at high altitude. The results indicate that this finding is independent of differences in BAD between the two populations, since it was confirmed by allometric scaling analysis, which has been recently indicated as an appropriate method taking into account for BAD differences between groups and for the non-linear relation between baseline BAD and its flow-mediated increase [22]. Furthermore endothelial dysfunction in HA occurs in the presence of preserved, or even enhanced, endothelium-independent vasodilation, thus excluding smooth muscle cell dysfunction as a cause of reduced FMD.

R.M. Bruno et al. / Atherosclerosis 236 (2014) 47e53

51

Fig. 2. Dot plots representing mean carotid diameter, carotid intima-media thickness (IMT) and circumferential wall stress in Caucasian sea-level controls (C, gray circles) and Himalayan high-altitude individuals (HA, black circles).

Interestingly, hyperemic FV, the stimulus for FMD in the brachial artery, is reduced in HA. The measurement of hyperemic stimulus, which is considered an index of microcirculatory dysfunction, has become mandatory during FMD exams, since recent studies suggest that its predictive role might be even greater than FMD itself, at least in younger populations [29,30]. Furthermore, hyperemic FV is regulated by endothelial NO release in the microcirculation [31]. Our results showed an attenuation of the difference in FMD between HA and C when hyperemic FV was included as a covariate, suggesting that reduced FMD in HA is at least in part a consequence of microcirculatory dysfunction. Noteworthy, reduced hyperemic FV was a relevant, independent determinant of FMD in HA, at variance to what observed in healthy Caucasian volunteers. The lack of correlation between FMD and hyperemic velocity in the C group is not surprising [32] and may be a consequence of biological variability in the FMD response to shear stress, attributable not only to the magnitude of the shear-stress stimulus, but also to the transduction of the vasodilator response to the smooth muscle cells, the response of the smooth muscle cells to a given vasodilator signal (which is influenced also by autonomic tone), and the structural characteristics of the vessel wall [33]. Such biological variability might be lost in HA under the pressure of an extreme environment. Alternatively, the FMD dependence from hyperemic FV in HA suggests that the reduced microcirculatory function may influence FMD in a dose-dependent manner in this specific population, at variance with Caucasian lowlanders. Another interesting feature of HA vascular phenotype is increased BA dilation to nitrates. This finding might be a consequence of a reduced autonomic restraint on vasculature [34,35], related to a parasympathetic predominance in the sympatho-vagal balance [36]. Acute exposure to high altitude causes an increase in heart rate and a shift of the sympatho-vagal balance towards increased sympathetic tone, but this response is blunted in Tibetans as compared to Han Chinese individuals [37]. The clinical significance of this presumably inherited condition is unknown at present. Pathophysiological mechanisms underlying the FMD reduction observed in HA are unknown. We suggest that the peculiar genetic background in Himalayan dwellers, protecting them from polycythemia and pulmonary hypertension, can negatively influence stimulated NO release. Mutations in the hypoxia-inducible factor (HIF) pathway were found by independent research groups in Tibetans and associated with their favorable phenotype [25,26]. Complex interrelationships between HIF and NO pathways have been documented [38], supporting this hypothesis, which however is highly speculative at the moment. Furthermore, hypoxia per se does not seem to play a relevant role, since acute O2 administration was not able to restore vascular function. An alternative explanation might also be hypothesized. Tonically increased NO

production, inducing chronic vasodilation, might have hampered stimulated NO release, although the confirmation of reduced BA diameter dilation in the allometric scaling analysis, the similar resting FV in the two groups, and the lack of dependence of FMD from BAD in HA argue against this hypothesis. To date, we do not know whether a reduced FMD secondary to “reduced NO-mediated vasodilating reserve” might have the same deleterious consequences on cardiovascular health than endothelial dysfunction due to reduced NO production or increased NO destruction. Which is the clinical significance of reduced FMD in Himalayan HA dwellers? The lack of relationship between systolic PAP or SO2 and FMD suggest that within this population a lower FMD is not simply an index of maladaptation to the challenging environment. Worth of note, in HA, as well as in C, a lower FMD is related to increased LV mass and carotid IMT, which are established surrogate endpoints of cardiovascular events. Thus, reduced FMD might be associated with increased cardiovascular risk even in this population. Although to date the prevalence of cardiovascular risk factors and disease in this area is unknown, ischemic heart disease and diabetes represent the first cause of death in developing countries, with their health burden increasing over years [39]. As far as vascular structure and geometry is concerned, we found no difference in aortic and peripheral PWV and in wave reflection between HA and C, excluding the presence of structural vascular alterations in these regions. However, since significant changes in PWV are expected only over 50 years, this parameter could not be enough sensitive in the relatively young population studied [40]. On the other hand, large artery geometry was profoundly altered: enlarged diameter, whose possible causes were already discussed, was accompanied by reduced IMT, leading to reduced W/L and increased circumferential wall stress. A reduced IMT might indicate a slower development of atherosclerosis and medial hypertrophy in HA as compared to C. Nevertheless, other mechanisms might be involved in the development of this peculiar carotid phenotype, although the cross-sectional design of this study limit the strength of this hypothesis. Some reports suggested that IMT can be considered a dynamic parameter, being acutely reduced when a nitrate is administrated to healthy subjects, as a feature of smooth muscle relaxation [41], or after a strenuous exercise in triathlon athletes [42]. In HA reduced IMT might be another expression of chronic vasodilation, together with enlarged carotid diameter. Another possible explanation is the occurrence of chronic remodeling, due to adaptation to extreme hemodynamic conditions. In chronic kidney disease, a condition characterized by chronic pressure/volume overload, a maladaptive remodeling occurs, with progressive, fast reduction of intima-media thickness and carotid enlargement, probably mediated by excessive extracellular matrix turnover, lack of vascular smooth muscle cell proliferation, or apoptosis [43]. The chronic hyperkinetic state due to

52

R.M. Bruno et al. / Atherosclerosis 236 (2014) 47e53

hypoxia might have caused in HA a carotid remodeling comparable to that observed in patients with chronic kidney disease. Furthermore, a genetic contribution cannot be excluded. Strengths of this study include a large sample size, a comprehensive multiparametric evaluation of cardiovascular function and a solid, standardized methodology. Limitations include the unavailability of blood samples, not authorized by the ethical committee, which did not allow us to investigate genetic differences, to consider renal function and metabolic profile among as risk factors and to ascertain the role of increased blood viscosity in HA. However, on the basis of current literature [1], we expect normal or only slightly increased hematocrit values in HA, with limited influence on shear stress and FMD. Other relevant limitations are lack of information about menstrual cycle phase, that could be a relevant confounder in a young population with a predominance of the female gender, and about the reproducibility of vascular function tests in the high altitude setting and population. Furthermore, in our study microcirculation is evaluated only with indirect methodologies, so the possible role of microcirculatory responses in influencing the observed alterations in large arteries cannot be completely taken into account for. Finally, data on flow velocity were not available in a relatively high number of participants, thus weakening our conclusions. In conclusion, this study demonstrated a unique vascular phenotype in Himalayan high altitude dwellers, characterized by large arteries dilation, a mainly microcirculatory endothelial dysfunction and a peculiar carotid remodeling as compared to Caucasian volunteers studied at the sea level. Reduced FMD occurs in the absence of classical cardiovascular risk factors and is related to surrogate endpoints for cardiovascular events. Prospective studies are needed to ascertain possible clinical consequences of the described vascular alterations in this population, which however present a favorable cardiac adaptation to chronic hypoxia, with a preserved systolic and diastolic left ventricular function and only modest increases in pulmonary pressure and resistance. The study of cardiovascular physiology in Himalayan dwellers might help to understand different routes of adaptation in populations chronically living at high altitude (more than 140 million people worldwide); furthermore, it might also constitute a model that can be translated to other settings, such as lowlanders going to high altitude for recreational or working purposes, and patients with chronic diseases characterized by hypoxia. Source of funding The study was supported by EV K2 CNR as part of the SHARE project. Disclosures None. Conflict of interest The authors report no relationships that could be construed as a conflict of interest. Acknowledgments None. Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.atherosclerosis.2014.06.017.

References [1] Beall CM. Two routes to functional adaptation: Tibetan and Andean highaltitude natives. Proc Natl Acad Sci U S A 2007;104(Suppl. 1):8655e60. [2] Chen QH, Ge RL, Wang XZ, Chen HX, Wu TY, Kobayashi T, et al. Exercise performance of tibetan and han adolescents at altitudes of 3,417 and 4,300 m. J Appl Physiol 1997;83:661e7. [3] Ge RL, Chen QH, Wang LH, Gen D, Yang P, Kubo K, et al. Higher exercise performance and lower vo2max in tibetan than han residents at 4,700 m altitude. J Appl Physiol 1994;77:684e91. [4] Groves BM, Droma T, Sutton JR, McCullough RG, McCullough RE, Zhuang J, et al. Minimal hypoxic pulmonary hypertension in normal tibetans at 3,658 m. J Appl Physiol 1993;74:312e8. [5] Ge RL, Witkowski S, Zhang Y, Alfrey C, Sivieri M, Karlsen T, et al. Determinants of erythropoietin release in response to short-term hypobaric hypoxia. J Appl Physiol 2002;92:2361e7. [6] Beall CM, Laskowski D, Strohl KP, Soria R, Villena M, Vargas E, et al. Pulmonary nitric oxide in mountain dwellers. Nature 2001;414:411e2. [7] Zhuang J, Droma T, Sun S, Janes C, McCullough RE, McCullough RG, et al. Hypoxic ventilatory responsiveness in tibetan compared with han residents of 3,658 m. J Appl Physiol 1993;74:303e11. [8] Bruno RM, Taddei S. Nitric oxide. In: Skinner FCMJS, editor. Encyclopedia of exercise medicine in health and disease. Berlin Heidelberg: Springer-Verlag; 2011. [9] Berger MM, Hesse C, Dehnert C, Siedler H, Kleinbongard P, Bardenheuer HJ, et al. Hypoxia impairs systemic endothelial function in individuals prone to high-altitude pulmonary edema. Am J Respir Crit Care Med 2005;172:763e7. [10] Goerre S, Wenk M, Bartsch P, Luscher TF, Niroomand F, Hohenhaus E, et al. Endothelin-1 in pulmonary hypertension associated with high-altitude exposure. Circulation 1995;91:359e64. [11] Bruno RM, Rossi L, Fabbrini M, Duranti E, Di Coscio E, Maestri M, et al. Renal vasodilating capacity and endothelial function are impaired in patients with obstructive sleep apnea syndrome and no traditional cardiovascular risk factors. J Hypertens 2013;31:1456e64. discussion 1464. [12] Lang RM, Bierig M, Devereux RB, Flachskampf FA, Foster E, Pellikka PA, et al. Recommendations for chamber quantification: a report from the American society of echocardiography's guidelines and standards committee and the chamber quantification writing group, developed in conjunction with the European association of echocardiography, a branch of the European society of cardiology. J Am Soc Echocardiogr Official Publ Am Soc Echocardiogr 2005;18:1440e63. [13] Quinones MA, Otto CM, Stoddard M, Waggoner A, Zoghbi WA. Recommendations for quantification of Doppler echocardiography: a report from the Doppler quantification task force of the nomenclature and standards committee of the American society of echocardiography. J Am Soc Echocardiogr Off Publ Am Soc Echocardiogr 2002;15:167e84. [14] Yock PG, Popp RL. Noninvasive estimation of right ventricular systolic pressure by Doppler ultrasound in patients with tricuspid regurgitation. Circulation 1984;70:657e62. [15] Nagueh SF, Middleton KJ, Kopelen HA, Zoghbi WA, Quinones MA. Doppler tissue imaging: a noninvasive technique for evaluation of left ventricular relaxation and estimation of filling pressures. J Am Coll Cardiol 1997;30: 1527e33. [16] Argiento P, Vanderpool RR, Mule M, Russo MG, D'Alto M, Bossone E, et al. Exercise stress echocardiography of the pulmonary circulation: limits of normal and sex differences. Chest 2012;142:1158e65. [17] Bruno RM, Penno G, Daniele G, Pucci L, Lucchesi D, Stea F, et al. Type 2 diabetes mellitus worsens arterial stiffness in hypertensive patients through endothelial dysfunction. Diabetologia 2012;55:1847e55. [18] Ghiadoni L, Faita F, Salvetti M, Cordiano C, Biggi A, Puato M, et al. Assessment of flow-mediated dilation reproducibility: a nationwide multicenter study. J Hypertens 2012;30:1399e405. [19] Giannarelli C, Bianchini E, Bruno RM, Magagna A, Landini L, Faita F, et al. Local carotid stiffness and intima-media thickness assessment by a novel ultrasound-based system in essential hypertension. Atherosclerosis 2012;223:372e7. [20] Boutouyrie P, Germain DP, Fiessinger JN, Laloux B, Perdu J, Laurent S. Increased carotid wall stress in vascular ehlers-danlos syndrome. Circulation 2004;109:1530e5. [21] Benetos A, Gautier S, Labat C, Salvi P, Valbusa F, Marino F, et al. Mortality and cardiovascular events are best predicted by low central/peripheral pulse pressure amplification but not by high blood pressure levels in elderly nursing home subjects: the partage (predictive values of blood pressure and arterial stiffness in institutionalized very aged population) study. J Am Coll Cardiol 2012;60:1503e11. [22] Atkinson G, Batterham AM. Allometric scaling of diameter change in the original flow-mediated dilation protocol. Atherosclerosis 2013;226:425e7. [23] Blitzer ML, Loh E, Roddy MA, Stamler JS, Creager MA. Endothelium-derived nitric oxide regulates systemic and pulmonary vascular resistance during acute hypoxia in humans. J Am Coll Cardiol 1996;28:591e6. [24] Blitzer ML, Lee SD, Creager MA. Endothelium-derived nitric oxide mediates hypoxic vasodilation of resistance vessels in humans. Am J Physiol 1996;271: H1182e5.

R.M. Bruno et al. / Atherosclerosis 236 (2014) 47e53 [25] Simonson TS, Yang Y, Huff CD, Yun H, Qin G, Witherspoon DJ, et al. Genetic evidence for high-altitude adaptation in tibet. Science 2010;329:72e5. [26] Yi X, Liang Y, Huerta-Sanchez E, Jin X, Cuo ZX, Pool JE, et al. Sequencing of 50 human exomes reveals adaptation to high altitude. Science 2010;329: 75e8. [27] Erzurum SC, Ghosh S, Janocha AJ, Xu W, Bauer S, Bryan NS, et al. Higher blood flow and circulating no products offset high-altitude hypoxia among tibetans. Proc Natl Acad Sci U S A 2007;104:17593e8. [28] Rimoldi SF, Rexhaj E, Pratali L, Bailey DM, Hutter D, Faita F, et al. Systemic vascular dysfunction in patients with chronic mountain sickness. Chest 2012;141:139e46. [29] Anderson TJ, Charbonneau F, Title LM, Buithieu J, Rose MS, Conradson H, et al. Microvascular function predicts cardiovascular events in primary prevention: long-term results from the firefighters and their endothelium (fate) study. Circulation 2011;123:163e9. [30] Ostad MA, Wild PS, Schnorbus B, Schulz A, Munzel T, Warnholtz A. Predictive value of brachial reactive hyperemia and flow-mediated dilation in stable coronary artery disease. Clin Hemorheol Microcirc 2014;56:247e57. [31] Meredith IT, Currie KE, Anderson TJ, Roddy MA, Ganz P, Creager MA. Postischemic vasodilation in human forearm is dependent on endotheliumderived nitric oxide. Am J Physiol 1996;270:H1435e40. [32] Thijssen DH, Bullens LM, van Bemmel MM, Dawson EA, Hopkins N, Tinken TM, et al. Does arterial shear explain the magnitude of flow-mediated dilation?: a comparison between young and older humans. Am J Physiol Heart Circ Physiol 2009;296:H57e64. [33] Thijssen DH, Black MA, Pyke KE, Padilla J, Atkinson G, Harris RA, et al. Assessment of flow-mediated dilation in humans: a methodological and physiological guideline. Am J Physiol Heart Circ Physiol 2011;300: H2e12.

53

[34] Santini L, Capria A, Brusca V, Violo A, Smurra F, Scarfo I, et al. An increased endothelial-independent vasodilation is the hallmark of the neurally mediated syncope. Clin Cardiol 2012;35:107e10. [35] Bruno RM, Sicari R, Corciu AI, Bianchini E, Faita F, Di Stefano R, et al. Noncancer atherosclerotic effects associated with environmental and therapeutic radiation doses: the chernobyl thyroid cancer children study. Int J Cardiol 2013;168:4255e7. [36] Zhuang J, Droma T, Sutton JR, McCullough RE, McCullough RG, Groves BM, et al. Autonomic regulation of heart rate response to exercise in tibetan and han residents of lhasa (3,658 m). J Appl Physiol 1993;75:1968e73. [37] Zhuang J, Zhu H, Zhou Z. Reserved higher vagal tone under acute hypoxia in tibetan adolescents with long-term migration to sea level. Jpn J Physiol 2002;52:51e6. [38] Ho JJ, Man HS, Marsden PA. Nitric oxide signaling in hypoxia. J Mol Med 2012;90:217e31. [39] Abegunde DO, Mathers CD, Adam T, Ortegon M, Strong K. The burden and costs of chronic diseases in low-income and middle-income countries. Lancet 2007;370:1929e38. [40] Determinants of pulse wave velocity in healthy people and in the presence of cardiovascular risk factors. Establishing normal and reference values. Eur Heart J 2010;31:2338e50. [41] Thijssen DH, Scholten RR, van den Munckhof IC, Benda N, Green DJ, Hopman MT. Acute change in vascular tone alters intima-media thickness. Hypertension 2011;58:240e6. [42] Bruno RM, Bianchini E, Vezzoli A, Mrakic Sposta S, Corciu A, Comassi M, et al. Acute changes in carotid diameter and thickness after an ironman competition. Eur Heart J 2013;34:1078. [43] Briet M, Collin C, Karras A, Laurent S, Bozec E, Jacquot C, et al. Arterial remodeling associates with ckd progression. J Am Soc Nephrol JASN 2011;22:967e74.