Preferential macrovasculopathy in systemic sclerosis detected by regional pulse wave velocity from wave intensity analysis: Comparisons of local and regional arterial stiffness parameters in cases and controls

Authors


Abstract

Objective

To study the extent and severity of macrovasculopathy in systemic sclerosis (SSc; scleroderma) patients by comparing both local and regional arterial stiffness parameters.

Methods

The local arterial stiffness indices of the right common carotid artery, right brachial artery, right radial artery, right superficial femoral artery, and right posterior tibial artery were measured in 25 SSc patients and strictly matched healthy controls. The regional pulse wave velocity (PWV) of each arterial segment was also calculated from wave intensity analysis.

Results

There were no differences between the two groups in the stiffness index (β), Peterson's pressure modulus, arterial compliance, and local PWV derived from β (PWVβ) of all vessels except the right brachial artery, of which β, Peterson's pressure modulus, and PWVβ were markedly lower and arterial compliance was higher in SSc patients compared with controls (P < 0.05). The forearm (brachial–radial) and arm (carotid–radial) PWVs were significantly higher in SSc patients than in controls (mean ± SD 12.1 ± 7.1 meters/second versus 8.3 ± 3.5 meters/second and mean ± SD 7.9 ± 1.9 meters/second versus 6.9 ± 1.5 meters/second, respectively; P < 0.05), whereas the upper arm (carotid–brachial), aortic (carotid–femoral), and leg (femoral–ankle) PWVs were not different between groups. The aortic PWV was also higher in the diffuse cutaneous SSc subgroup than in controls (mean 6.2, 95% confidence interval [95% CI] 5.4–6.9 meters/second versus mean 5.1, 95% CI 4.7–5.6 meters/second; P < 0.05) after adjusting for potentially influential variables.

Conclusion

The macrovasculopathy occurs preferentially at the forearm and aorta in SSc, which can be sensitively and reliably detected by regional PWVs rather than commonly used local arterial stiffness indices.

INTRODUCTION

Systemic sclerosis (SSc; scleroderma) is a generalized connective tissue disorder of unknown etiology characterized by thickening and fibrosis of the skin and distinctive visceral involvement associated with vascular damage (1). Its most typical early manifestation is Raynaud's phenomenon (RP), which occurs in more than 90% of patients with SSc (2). Traditionally, the vasculopathy of SSc has been considered mainly to affect small arteries and capillaries (3); however, there is recent evidence showing that SSc is associated with the prevalence of large vessel disease (4), endothelial dysfunction, and increased arterial wall stiffness (5). Although a number of studies were available regarding the increasing arterial stiffness of SSc patients in general, their results still remained discrepant in detail (6); furthermore, most of their parameters used for analyses were only restricted to the stiffness index (β) of the common carotid artery, the aortic augmentation index, and the carotid–femoral pulse wave velocity (PWV), all of which mainly reflect the stiffness of large elastic arteries. In contrast, the stiffness of distal medium muscular arteries has not yet been well investigated. Additionally, the relationships between arterial stiffness parameters and the macrovascular stiffening per se in SSc patients still remain to be fully explored.

Recently, wave intensity analysis has become a useful method for evaluating traveling pressure and flow waves in the aorta and larger arteries. Some in-depth overviews of wave intensity analysis and its application were published previously (7, 8). The wave intensity and arterial stiffness indices now can be obtained simultaneously and immediately through the wave intensity analysis ultrasound system, which is based on the Doppler and echo-tracking technique. Using this novel technique, the present study was designed to investigate the systemic macrovascular impairment in a Chinese SSc cohort, which has rarely been performed before, by examining the arterial stiffness parameters of both central elastic and peripheral muscular arteries. This study therefore offered an opportunity to test which arterial stiffness parameters are more suitable for detecting the early macrovascular stiffening per se in SSc patients and which arteries are apt to be involved in the macrovasculopathy.

PATIENTS AND METHODS

Patients and controls.

Twenty-five consecutive patients with an established SSc diagnosis according to the American College of Rheumatology criteria (9) were recruited from the Department of Rheumatology in Tangdu Hospital for the investigation, which had been approved by the ethics committee of the hospital and conformed to the principles of the Declaration of Helsinki. The disease duration was defined based on the onset of RP. Twenty-five strictly matched healthy subjects were randomly selected as controls. All of the participants, including both patients and controls, were of the same ethnicity, i.e., Han nationality. None of the participants were smokers or had evidence of concomitant hypertension, arrhythmia, diabetes mellitus, dyslipidemia, known cardiovascular disease, vasculitis, and infection.

Measurements.

All of the measurements were performed by a single trained sonographer (JL) unaware of the participants' status in the morning after they had fasted overnight and abstained from alcohol and caffeine. Participants rested supine in a quiet, air-conditioned room (22–24°C) for at least 10 minutes before undergoing the following measurements. Blood pressure was measured 3 times in 5-minute intervals by an electronic sphygmomanometer (Omron HEM-7052) at the right upper arm, which was also the dominant arm, and the mean value was used in the study. Five arteries, including the right common carotid artery (RCCA), right brachial artery (RBA), right radial artery (RRA), right superficial femoral artery (RSFA), and right posterior tibial artery (RPTA), were studied at the following sites: 3 cm proximal to the bifurcation of the common carotid artery, 3 cm above the antecubital fossa, 3 cm above the wrist, 3 cm distal to the bifurcation of the common femoral artery, and 3 cm above the medial ankle, respectively. In addition, the intima-media thickness (IMT) of the posterior wall of each artery was determined by B-mode ultrasonography of high resolution at the sampling site. All of the arteries were examined thoroughly at full length in order to exclude regional hemodynamically relevant stenosis. Participants with such stenosis were not found in the study.

Wave intensity and local arterial stiffness indices.

This wave intensity analysis system is incorporated in ultrasonic diagnostic equipment (ProSound α10), which has a color Doppler system for blood flow velocity measurements and an echo-tracking subsystem for diameter change measurements with a 5–13 MHz linear array transducer. The details of this system were described elsewhere (10). Briefly, after setting the tracking positions, displayed as two small dotted bars along the anterior and posterior arterial walls perpendicular to the ultrasound beam line (line a in Figure 1A), the echo-tracking subsystem automatically starts measurements of arterial diameter change. The blood flow velocity averaged along the Doppler beam (line b in Figure 1A) crossing the artery is measured using range-gated color Doppler signals. At least 5 consecutive beats were ensemble averaged to obtain a representative waveform. After the measured blood pressure data were inputted for calibration, the wave intensity and local arterial stiffness indices were calculated automatically and displayed on the monitor (Figure 1B).

Figure 1.

A, simultaneous measurements of arterial diameter change and blood flow velocity in color Doppler/B (left) and M (right) modes, B, the output display of wave intensity (WI) and arterial stiffness indices, and C, calculation of regional pulse wave velocity (PWV) from WI and synchronous electrocardiogram waveform. The carotid–femoral transit distance (cfTD) was estimated by taking the difference of the distance from the suprasternal notch (SSN) to the carotid site (SSN-C) and the distance from the SSN to the femoral site (SSN-F) into account for parallel transmission along the brachiocephalic and carotid arteries and around the aortic arch (red shading). Note that the carotid–femoral PWV (cfPWV) and carotid–brachial PWV fails to assess the stiffness of the proximal aorta (red shading). The carotid–femoral transit time (cfTT) was calculated by subtracting the interval between the R wave of electrocardiogram and the first peak of WI (R − W1) of the right common carotid artery (RCCA) from that of the right superficial femoral artery (RSFA), and a similar procedure was used to compute other TTs. RBA = right brachial artery; RRA = right radial artery; RPTA = right posterior tibial artery; § = SSN; br = brachial–radial; fa = femoral–ankle. β = stiffness index; Ep = Peterson's pressure modulus; AC = arterial compliance; PWVβ = local PWV derived from β; W1 = first peak of WI; W2 = second peak of WI; NA = negative area between W1 and W2; R-1st = R − W1; 1st-2nd = W1 − W2.

Wave intensity is computed as (dP/dt) × (dU/dt), where dP and dU = changes in blood pressure and velocity, respectively, during constant short time intervals (dt). The wave intensity indices here include two intensive indices (the first peak [W1] and the second peak [W2]), two temporal indices (the interval between the R wave of electrocardiogram and W1 [R − W1] and the interval between W1 and W2 [W1 − W2]), and the negative area between W1 and W2 (NA), which indicates the effects of reflected waves.

The local arterial stiffness indices, including β, Peterson's pressure modulus, arterial compliance, and local PWV derived from β (PWVβ), can be calculated by the echo-tracking subsystem according to the following formulas: β = ln(Ps/Pd)/[(Ds − Dd)/Dd], Peterson's pressure modulus = (Ps − Pd)/[(Ds − Dd)/Dd], arterial compliance = π(Ds × Ds − Dd × Dd)/[4 (Ps − Pd)], and PWVβ = equation image(11), where Ps = end systolic pressure; Pd = end diastolic pressure; Ds = maximum arterial diameter and Dd = minimum arterial diameter, measured by wall tracking of the medial–adventitial borders of the arteries; and ρ = blood density (ρ = 1,050 kg × m−3).

Regional arterial stiffness parameter: regional PWV.

Regional PWV was calculated for each arterial segment as the pulse wave transit distance (TD) between two sampling sites divided by the corresponding pulse wave transit time (TT) delay: PWV = TD/TT (meters/second). The aortic (carotid–femoral), upper arm (carotid–brachial), and arm (carotid–radial) TDs were measured by subtraction method (12), while the leg (femoral–ankle) and forearm (brachial–radial) TDs were measured by direct site-to-site method, i.e., superficial measurement of the distance between the two sampling sites. The TT on the arterial segment was calculated by subtracting R − W1 of the proximal artery from that of the distal artery (Figure 1C).

Intraobserver reproducibility for repeated measurements of arterial stiffness parameters.

Fifteen subjects from the control group were randomly selected for assessment of the intraobserver intersession variability in repeated measurements of each arterial stiffness parameter. Both measurements were performed in the same condition with an interval of 1 day. The variability was calculated using Bland-Altman analyses (13) and expressed as percentages of 95% confidence limits of the difference (calculated as equation image SDs) from the mean value of the paired measurements (10).

Statistical analysis.

Data are expressed as the mean ± SD for normally distributed continuous variables (including age, disease duration, diameter, IMT, R − W1, β, Peterson's pressure modulus, arterial compliance, PWVβ, and regional PWVs) or as the median (interquartile range) for non–normally distributed continuous variables (including creatinine level and Medsger severity score), unless otherwise indicated. Differences between variables that assumed normal distributions were investigated by means of an unpaired t-test, while variables that did not assume normal distributions were investigated using the nonparametric Mann-Whitney U test. Pearson's correlation was used to investigate the correlations between the normally distributed variables. A general linear model was applied to estimate differences in means of the regional PWVs between the SSc subgroups and control group after adjustment for possible confounding factors. All of the data were analyzed using the statistical software SPSS, version 15.0. For all tests, P values less than 0.05 (2-tailed) were considered statistically significant.

RESULTS

Characteristics of the study population.

All of the patients (mean ± SD disease duration 4.9 ± 5.6 years) had RP, and only 3 (12%) had digital ulcers. Seventeen patients (68%) had limited cutaneous systemic sclerosis (lcSSc) and the other 8 (32%) had diffuse cutaneous systemic sclerosis (dcSSc) (14). Twenty-one patients (84%) had pulmonary fibrosis or pulmonary arterial hypertension, where pulmonary artery systolic pressures of >35 mm Hg on echocardiogram were used as a screening cutoff point. Cardiac abnormalities were observed in 12 patients (48%). Nine patients (36%) had gastrointestinal and 2 patients (8%) had renal manifestations. Four patients (16%) had musculoskeletal involvement and 3 patients (12%) had Sicca symptoms. Common characteristics of the SSc patients and healthy controls as well as the Medsger severity score (15, 16) and pulmonary arterial hypertension degree of the SSc patients are described in Table 1. The two groups were well matched with respect to age, sex, size, body mass index (BMI), blood pressure, left ventricular ejection fraction, and metabolic profile. Regarding SSc patients' current therapy within 24 hours, the drugs used were: proton-pump inhibitors (80%), steroids (32%), immunosuppressive agents (16%), alfacalcidol (48%), nonsteroidal antiinflammatory drugs (12%), diuretics (8%), penicillamine (12%), aspirin (68%), dipyridamole (64%), beta-blockers (8%), angiotensin-converting enzyme (ACE) inhibitors or angiotensin II receptor blockers (8%), calcium-channel blockers (12%), as well as traditional Chinese medicine, including sodium tanshinone IIA sulfonate (STS; 84%) and scutellarin (60%).

Table 1. Demographic characteristics of the SSc patients and healthy controls*
 Controls (n = 25)SSc (n = 25)P
  • *

    Values are the mean ± SD unless otherwise indicated. 1 mm Hg = 0.133 kPa. SSc = systemic sclerosis; NS = not significant; BMI = body mass index; BP = blood pressure; LVEF = left ventricular ejection fraction; IQR = interquartile range; PASP = pulmonary artery systolic pressure.

  • Two patients with SSc, but no controls, had a serum creatinine level above 117 μmoles/liter.

Age, years46.2 ± 11.247.2 ± 10.1NS
Sex, male/female3/223/22NS
Height, cm159.5 ± 5.8158.6 ± 5.7NS
Weight, kg54.6 ± 5.655.3 ± 7.4NS
BMI, kg/m221.45 ± 1.9221.98 ± 2.61NS
Heart rate, beats/minute71.6 ± 8.475.0 ± 14.9NS
Systolic BP, mm Hg117.6 ± 12.7112.9 ± 11.3NS
Diastolic BP, mm Hg72.1 ± 7.769.8 ± 6.2NS
Pulse pressure, mm Hg45.5 ± 6.343.1 ± 7.4NS
LVEF, %66 ± 768 ± 5NS
Plasma glucose, mmoles/liter4.62 ± 0.394.51 ± 0.59NS
Creatinine, median (IQR) μmoles/liter57.0 (49.5–63.5)60.0 (53.0–65.0)NS
Total cholesterol, mmoles/liter4.48 ± 0.604.33 ± 0.73NS
Triglycerides, mmoles/liter0.99 ± 0.280.91 ± 0.39NS
Medsger severity scale score, median (IQR)07.0 (5.0–9.5) 
Echo PASP, no.   
 <35 mm Hg254 
 35–65 mm Hg018 
 >65 mm Hg03 

Diameter, IMT, R − W1, and local arterial stiffness indices.

No significant differences were found between the two groups in terms of the vessel diameter, IMT, and R − W1, except for the R − W1 of the RRA, which was markedly shortened in the SSc patients compared with the controls (mean ± SD 162.6 ± 20.1 msec versus 175.2 ± 17.6 msec; P < 0.05), as shown in Table 2. Because the other wave intensity indices, including W1, W2, W1 − W2, and NA, were not related to the arterial stiffness or were reported to lack reproducibility (e.g., W2 and NA) (10), the data for them were not included here for discussion, although they also showed no differences between the two groups. There were no significant differences between the two groups in the local arterial stiffness indices of all arteries except the RBA, of which the β, Peterson's pressure modulus, and PWVβ were markedly lower and the arterial compliance was apparently higher in SSc patients than in healthy controls (P < 0.05).

Table 2. Diameter, IMT, R − W1, and local arterial stiffness indices in the SSc and control groups*
VesselDiameter, mmIMT, mmR − W1, msecβPeterson's pressure modulus, kPaArterial compliance, mm2/kPaPWVβ, meters/second
ControlSScControlSScControlSScControlSScControlSScControlSScControlSSc
  • *

    Values are the mean ± SD. IMT = intima-media thickness; R − W1 = interval between the R wave of electrocardiogram and the first peak; SSc = systemic sclerosis; β = stiffness index; PWVβ = local pulse wave velocity derived from β; RCCA = right common carotid artery; RBA = right brachial artery; RRA = right radial artery; RSFA = right superficial femoral artery; RPTA = right posterior tibial artery.

  • Diameter here indicates the maximum diameter that was automatically measured by echo tracking and does not refer to the real maximum inner diameter but an approximation of the maximum inner diameter plus the posterior and anterior IMT of the examined vessel.

  • P < 0.01 versus the control group.

  • §

    P < 0.05 versus the control group.

RCCA7.35 ± 0.917.40 ± 0.610.447 ± 0.0860.466 ± 0.06999.5 ± 11.296.6 ± 22.78.93 ± 2.029.44 ± 3.42111.6 ± 32.3113.4 ± 44.80.744 ± 0.2080.775 ± 0.2256.36 ± 0.916.38 ± 1.15
RBA3.92 ± 0.563.92 ± 0.600.280 ± 0.0370.286 ± 0.033144.5 ± 12.7139.1 ± 21.821.42 ± 6.9015.80 ± 6.03265.3 ± 91.5191.4 ± 81.60.098 ± 0.0440.159 ± 0.098§9.76 ± 1.758.22 ± 1.84
RRA2.27 ± 0.392.32 ± 0.490.223 ± 0.0290.226 ± 0.026175.2 ± 17.6162.6 ± 20.1§25.83 ± 8.9530.80 ± 13.94320.3 ± 115.9364.4 ± 167.10.026 ± 0.0110.034 ± 0.03910.68 ± 1.8711.24 ± 2.94
RSFA5.39 ± 0.635.62 ± 0.710.419 ± 0.0260.420 ± 0.054197.5 ± 27.8186.3 ± 28.915.67 ± 5.9417.88 ± 4.99194.3 ± 78.6211.2 ± 56.60.260 ± 0.1110.264 ± 0.1868.32 ± 1.688.73 ± 1.29
RPTA2.12 ± 0.392.01 ± 0.340.258 ± 0.0340.260 ± 0.041256.0 ± 24.5244.0 ± 27.522.88 ± 6.4026.03 ± 8.74284.2 ± 88.1310.8 ± 111.30.025 ± 0.0080.023 ± 0.01210.12 ± 1.6110.55 ± 1.98

Regional PWVs.

The forearm and arm PWVs in SSc patients were significantly higher than that observed in the controls (mean ± SD 12.1 ± 7.1 meters/second versus 8.3 ± 3.5 meters/second and 7.9 ± 1.9 meters/second versus 6.9 ± 1.5 meters/second, respectively; P < 0.05), whereas the upper arm, aortic, and leg PWVs showed no difference between the SSc and control groups (mean ± SD 6.7 ± 1.8 meters/second versus 6.5 ± 2.1 meters/second, 5.6 ± 1.3 meters/second versus 5.2 ± 1.3 meters/second, and 11.7 ± 4.1 meters/second versus 11.4 ± 3.7 meters/second, respectively; P > 0.05), as shown in Figure 2. The subgroup analyses showed that in addition to the forearm and arm PWVs, the aortic PWV was also higher in dcSSc patients as compared with controls after adjustment for age, sex, BMI, heart rate, and systolic and diastolic blood pressure; however, none of the regional PWVs showed statistical differences between the lcSSc subgroup and control group before and after adjustment (Table 3).

Figure 2.

Regional pulse wave velocity in the systemic sclerosis (SSc; n = 25) and control (n = 25) groups. Values are the mean ± SD. Forearm = brachial–radial; upper arm = carotid–brachial; arm = carotid–radial; aortic = carotid–femoral; leg = femoral–ankle; m/s = meters/second; ** = P < 0.01 versus the control group; NS = not significant versus the control group; * = P < 0.05 versus the control group.

Table 3. Regional PWVs of the SSc subgroups versus the control group after adjustment for age, sex, body mass index, heart rate, and systolic and diastolic blood pressure*
 Control (n = 25)lcSSc (n = 17)dcSSc (n = 8)
  • *

    Values are the mean (95% confidence interval). PWV = pulse wave velocity; SSc = systemic sclerosis; lcSSc = limited cutaneous SSc; dcSSc = diffuse cutaneous SSc; forearm = brachial–radial; upper arm = carotid–brachial; arm = carotid–radial; aortic = carotid–femoral; leg = femoral–ankle.

  • P < 0.01 versus the control group.

  • P < 0.05 versus the control group.

Forearm PWV, meters/second9.0 (7.0–10.9)8.9 (6.6–11.3)16.5 (13.0–19.9)
Upper arm PWV, meters/second6.5 (5.7–7.2)6.5 (5.5–7.4)7.3 (5.9–8.7)
Arm PWV, meters/second7.0 (6.4–7.6)7.1 (6.4–7.8)9.1 (8.0–10.2)
Aortic PWV, meters/second5.1 (4.7–5.6)5.4 (4.9–5.9)6.2 (5.4–6.9)
Leg PWV, meters/second12.1 (10.6–13.6)11.8 (10.1–13.6)9.2 (6.6–11.8)

Correlations of arterial stiffness parameters with age and disease duration.

Only the β, Peterson's pressure modulus, PWVβ of the RCCA, and aortic PWV displayed significant positive correlations to age in the control group, with the aortic PWV being of the highest correlation coefficient (r = 0.733, P < 0.001), as shown in Table 4. No significant correlations were found between the measured arterial stiffness parameters and disease duration in the SSc group (data not shown).

Table 4. Age-correlated coefficients (n = 25) and intraobserver intersession variabilities (n = 15) of arterial stiffness parameters in the control group*
VesselβPeterson's pressure modulus, kPaArterial compliance, mm2/kPaPWVβ, meters/secondRegional PWVs, meters/second
rV, %rV, %rV, %rV, %rV, %
  • *

    β = stiffness index; PWVβ = local pulse wave velocity derived from β; PWV = pulse wave velocity; r = correlation coefficient to age; V = intraobserver intersession variability; RCCA = right common carotid artery; RBA = right brachial artery; RRA = right radial artery; RSFA = right superficial femoral artery; RPTA = right posterior tibial artery; forearm = brachial–radial; upper arm = carotid–brachial; arm = carotid–radial; aortic = carotid–femoral; leg = femoral–ankle.

  • P < 0.05.

  • P < 0.01.

  • §

    P < 0.001.

RCCA0.42915.40.52516.6−0.1828.90.5047.3  
RBA−0.13723.60.04523.50.08621.60.02511.1  
RRA−0.06119.80.24819.90.08623.30.20410.0  
RSFA−0.07413.70.09413.40.0328.40.0966.4  
RPTA0.02622.30.19422.40.12532.30.15011.1  
Forearm PWV        0.2139.3
Upper arm PWV        0.1847.0
Arm PWV        0.2723.4
Aortic PWV        0.733§3.2
Leg PWV        −0.2715.9

Intraobserver reproducibility.

The intraobserver intersession variabilities were apparently lower in regional PWVs than in most of the local arterial stiffness parameters. The variabilities of local and regional arterial stiffness parameters appeared to be reversely related to the caliber of the sampling arteries and the distance of the arterial segments, respectively (Table 4), due to a larger variation in tone of smaller muscular arteries, which are under permanent neurohumoral control.

DISCUSSION

The IMTs in SSc patients have been studied by several researchers. Mourad et al (17) found that the radial artery internal diameter of SSc patients was significantly decreased, whereas the IMT and mean arterial pressure were closely similar to the controls. The carotid, femoral, and brachial IMTs in SSc patients also showed no differences from the controls (18–20). In accordance with previous studies, the IMTs measured at 5 different arteries in SSc patients showed no differences from the controls in the present study (Table 2). The mean value of the common carotid IMT here was somewhat lower than those reported previously, probably due to the ethnic or other differences in the studied populations. This finding confirmed that SSc is not associated with an increased prevalence of early signs of atherosclerosis.

Four different local arterial stiffness indices (β, Peterson's pressure modulus, arterial compliance, and PWVβ) were selected for analyses in the study to avoid bias toward any of the parameters. Curiously, only the local arterial stiffness indices of the RBA in SSc patients showed differences compared with controls, which appeared to indicate a decrease in brachial arterial stiffness of SSc patients. However, as we know, the stiffness of most arteries in SSc patients has been demonstrated to be increased in many reports, without any arteries in these patients reported to have decreased stiffness. The elastic properties of the carotid artery were reported to be decreased in SSc patients, whereas those of the femoral artery did not change (18). Andersen et al (21) found that radial artery wall stiffness was significantly greater in SSc patients than in controls. The stiffness of the brachial artery was seldom studied before, whereas its endothelial function was evaluated by a number of researchers but still with conflicting reports. Most of the reports indicated that the flow-mediated dilation (FMD%) was impaired while the nitroglycerin-mediated dilation was often preserved (22–24), but there were still at least two reports showing that the FMD% was also preserved in SSc (19, 21). The whole large-vessel stiffness in patients with SSc was also proven to be worsened by increasing in the augmentation index and PWV of the aorta (25). So the decreasing in local arterial stiffness indices of the RBA observed here could not be explained except for taking the effects of current medications of SSc patients into consideration. At the time of this study, vasoactive drugs that could potentially influence arterial stiffness included aspirin (26), dipyridamole, beta-blockers, ACE inhibitors or angiotensin II receptor blockers, calcium-channel blockers, as well as traditional Chinese medicine (STS and scutellarin). Pharmacologic studies indicate that the effects of long-term treatment with ACE inhibitors, calcium-channel antagonists, and some beta-blockers on arterial stiffness are generally similar (27), and they can all improve the arterial stiffening (28, 29). The STS (30) and scutellarin (31) have also been demonstrated to have vasorelaxant effects. These vasoactive drugs have little direct effect on large central elastic arteries, but can markedly change the PWV from the periphery to the heart by altering the stiffness of the peripheral medium-sized muscular arteries, which is modulated by the vasomotor tone, either depending on the endothelial function or sympathetic nervous system (32) or the renin–angiotensin system (33). Because most of the prescribed drugs for SSc patients had vasodilatory effects, it was also explicable that the local arterial stiffness indices of the other 4 arteries (RCCA, RRA, RSFA, and RPTA) in SSc patients showed no statistical differences as compared to controls, although with a tendency to increase arterial stiffness in these arteries. The results also implied that the brachial artery might be less damaged by SSc or that its local arterial stiffness indices are affected more by pharmacologic treatments than by the other arteries (34).

On the other hand, it must be admitted that random variation in the measurements of small muscular arteries was rather large (Table 4). Moreover, histologic work has described the structural differences between elastic and muscular arteries (35), and the noninvasive work further confirmed that the elastic moduli of the brachial and radial arteries are different from that of the common carotid artery (36). Therefore, the local arterial stiffness indices are not appropriate for evaluating the stiffness of a relatively small muscular artery, such as the radial or tibial artery, when considering the accuracy and reproducibility. Therefore, we took the regional PWVs for further analyses on segmental arterial stiffness. In our study, the forearm PWV increased apparently in the SSc group, and furthermore, the aortic PWV was also shown to be increased in the dcSSc subgroup as compared with the control group. The difference between the results of local and regional stiffness parameters may be due to the different influence of vasoactive therapy, which is going to be our next study. The normal values of regional PWVs in the study appeared to be lower than those in the previous reports (37), probably due to the age, ethnic, or other differences in the studied populations and different methods used to measure the regional PWVs.

In order to evaluate the validity of the abovementioned arterial stiffness parameters in reflecting the arterial stiffness per se, these parameters were made correlative analyses with age in the control group, because age is identified as a significant independent determinant of stiffness in elastic arteries. Among these parameters, only the β, Peterson's pressure modulus, PWVβ of the RCCA, and aortic PWV showed significant positive correlations to age, while only the RCCA is the elastic artery among the investigated vessels in this study. This result conformed to the previous reports showing that the stiffness parameters of the peripheral muscular arteries are modified appreciably by vasoactive stimuli, and the mechanical properties of the deeper elastic arteries provide sufficiently reliable information about changes caused by aging (38). Age-related stiffening affects predominantly the aorta and proximal elastic arteries, and to a lesser degree the peripheral medium-sized muscular arteries (39).

Among the arterial stiffness parameters that could detect the age-related changes in the stiffness of elastic arteries caused by medial calcification and loss of elasticity, the aortic PWV had the highest correlation with age in the study; moreover, only the regional PWVs other than the corresponding local arterial stiffness indices showed abnormality due to the macroangiopathy in SSc patents, which also implied that the local arterial stiffness indices may be more susceptible to vasoactive drugs than regional PWVs. Both results here confirmed that the regional PWV is a powerful independent indicator (gold standard index) of the arterial stiffening in both elastic and muscular arteries, and can be applied to detect the pathology prior to the appearance of morphologic changes of the vasculature (40). With no differences found in the caliber and IMT of each artery between the SSc and control groups (Table 2), we conclude that this regional PWV method can easily quantify the alteration of arterial wall mechanics, perhaps due to the fibrosis of the arterial media and adventitia reflected by the increased integrated backscatter values (41), which precede the morphologic and geometric changes under high-resolution B-mode ultrasound in SSc patients.

Taken together, this study suggests that the stiffening initially and predominantly affects the muscular arteries of the forearm in SSc patients, and as the disease develops it will be involved in the aorta and the proximal elastic arteries, whereas the muscular arteries of the leg and upper arm are less commonly affected. The result is in line with some epidemiologic evidence and previous reports that the radial artery wall stiffness was significantly greater in SSc patients than in controls (21), and the findings of macrovasculopathy in the ulnar and radial arteries of the SSc patients were also demonstrated by angiography (42). Macrovascular occlusion proximal to the digits seemed to be more frequent in the upper extremity than in the lower extremity (43). Therefore, patients with SSc may have preferential targets for the macrovasculopathy just as their skin involvements are of heterogeneity. We speculate that the preferential stiffening in muscular arteries of the forearm may be related to the high incidence of RP in the hands, which might lead inflammatory injury to their adjacent arteries, while the stiffening in elastic arteries might be relevant to the alteration of fibrillin 1 metabolism in SSc patients (44). The underlying mechanisms and the clinical implications of this phenomenon should be studied further.

To our knowledge, this is the first study to examine segmental macrovasculopathy throughout the entire arterial tree except cerebral arteries in patients with SSc, and to provide evidence of heterogeneity in their macrovascular complications. The major strength of our study is a design that allowed the concomitant measurements of both local and regional arterial stiffness parameters in a simple, objective, and time-saving way. Our study also had some limitations. First of all, ideally, the blood pressure used for calculating local arterial stiffness indices should be measured at the corresponding artery; however, we chose to use the external brachial artery pressure as the source because it was more accessible than other arteries, which would hinder precise measurements of the indices at these arteries, as mentioned previously. Second, the status of SSc patients (most of them were women) was heterogeneous with regard to menstrual cycle (45, 46), menopause status (47), and type of ongoing medical treatment (48), all of which were shown to affect the arterial stiffness parameters. Finally, the small sample size did not permit multivariant analyses for comparing arterial stiffness parameters in SSc patients with a different status.

In summary, the macrovasculopathy of SSc patients, characterized by arterial stiffening with no significant change in IMT, predominantly affects the muscular arteries of the forearm and the proximal elastic arteries, whereas the muscular arteries of the leg and upper arm are less commonly affected.

Regional PWV is a sensitive and reliable noninvasive index for quantifying the extent and degree of macrovascular stiffening in SSc patients, which might be occult in local arterial stiffness indices due to the pharmacologic effects or other influential factors in the clinic setting. The regional PWV autocalculation function should be integrated into the ultrasound system for routine screening of segmental macrovasculopathy. The medication withdrawal is ethically unacceptable in clinics, which inevitably introduces potential sources of error for examining arterial stiffness. Therefore, large-scale clinical trials are needed to examine the therapeutic intervention on the arterial stiffness parameters during the management of SSc. Further studies are needed to clarify whether the macrovascular abnormalities indicated by regional PWVs are correlated with skin sclerosis or internal organ manifestations and have an impact on the prognosis and treatment strategy. The macrovascular involvement stratification in SSc patients might act as a valuable guide in assessing the efficacy of therapeutic interventions in addition to the local microcirculation and blood pressure.

AUTHOR CONTRIBUTIONS

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Cao had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Liu, Cao.

Acquisition of data. Liu, Zhang, Yang, Li, Yao.

Analysis and interpretation of data. Liu, Cao, Duan, Yuan, Yang.

Ancillary