Silent cardiovascular involvement in patients with diffuse systemic sclerosis: a controlled cross-sectional study




An association between systemic autoimmune diseases and atherosclerosis has been described in many connective tissue diseases, and this association is known to lead to increased cardiovascular morbidity and mortality. Systemic sclerosis (SSc) is characterized by multisystem organ inflammation, endothelial wall damage, and vasculopathy. There are many markers of endothelial dysfunction and/or atherosclerotic risk, such as asymmetric dimethylarginine (ADMA), arterial stiffness parameters, carotid intima-media thickness (CIMT), and coronary flow reserve (CFR) assessed by transthoracic echocardiography. The aim of this pilot study was to use various endothelial and atherosclerosis markers to identify early cardiovascular involvement in a group of SSc patients.


The study involved 20 patients (2 men and 18 women with a mean ± SD age of 52.96 ± 12.51 years) with diffuse SSc who had no signs or symptoms of cardiovascular disease (CVD) and 20 age- and sex-matched controls. All subjects underwent a dipyridamole echocardiographic stress test that included a determination of CFR and an evaluation of CIMT, arterial stiffness, and plasma ADMA levels.


All of the arterial wall measurements of the patients with diffuse SSc were significantly different from those of the controls, and both right and left CIMT, pulse wave velocity, and stiffness index (β) were significantly elevated in the SSc patients compared to the healthy controls. Moreover, in patients with diffuse SSc, CFR was significantly lower (P = 0.0033) and plasma ADMA levels were higher (P < 0.0001) than in healthy controls.


SSc patients without any clinical evidence of CVD seem to have had subclinical atherosclerosis, which was suggested by early impairment of coronary microcirculation and macrovascular involvement.


Systemic sclerosis (SSc) is a multisystem disease characterized by vasculopathy, organ fibrosis, and endothelial cell injury, which may increase the risk of coronary atherosclerosis. It is thought that the vascular involvement is the result of an immune/inflammatory response that activates and injures the vascular endothelium (1), and it is known that inflammation and endothelial dysfunction are key mechanisms in the pathogenesis of atherosclerosis (2). The vasculopathy of SSc typically affects the small arteries and capillaries, although macrovascular disease has been demonstrated by carotid ultrasonography, the ankle brachial blood pressure index, and peripheral angiography (3, 4).

Coronary vessel involvement has been ascertained invasively by coronarography in a few small studies (4, 5), and also by simpler noninvasive methods, such as transthoracic echocardiography and the evaluation of coronary flow reserve (CFR). A number of studies have established the usefulness of CFR in the detection of coronary macrocirculation and microcirculation disorders in systemic autoimmune diseases (SADs) (6, 7), and many classic and novel cardiovascular (CV) indices have been formulated for the detection of subclinical atherosclerosis, including carotid intima-media thickness (CIMT) (8, 9). In patients with SADs, the most important parameters for evaluating endothelial dysfunction and early atherosclerosis are arterial distensibility and stiffness (10, 11). Arterial stiffness is a novel marker of vascular dysfunction that has recently been recognized as an independent risk factor of CV disease (CVD) (12), and Laurent et al (13) have evaluated arterial stiffness using pulse wave velocity (PWV).

It has also been suggested that asymmetric dimethylarginine (ADMA) may be another CV risk factor involved in endothelial dysfunction (14). ADMA is the major endogenous inhibitor of nitric oxide (NO) synthase that is known to be an indicator of cardiovasculopathy in many diseases, particularly in chronic inflammatory disorders (15, 16). The clinical relevance of vascular inflammation is unclear, although vascular inflammation does contribute to the atherosclerotic process. Moreover, traditional CV risk factors cannot explain the increased and accelerated atherosclerosis detectable in SADs, which is why many authors have attempted to identify the early markers of atherosclerosis (17). The aim of this pilot study was to identify early CV involvement in a group of patients with diffuse SSc using various endothelial or atherosclerosis markers.

Significance & Innovations

  • We used various endothelial and atherosclerosis markers to identify early cardiovascular involvement in a group of systemic sclerosis (SSc) patients.

  • In this study, SSc patients without any clinical evidence of cardiovascular disease seemed to have had subclinical atherosclerosis, which was suggested by early impairment of coronary microcirculation and macrovascular involvement.


Study population

The CV risk profile for 20 consecutive outpatients selected from a source cohort of 48 patients who fulfilled the American College of Rheumatology criteria for SSc (18, 19) and had no clinical history or signs of coronary artery disease, arrhythmia, or other cardiac diseases was assessed between May and November 2010; 2 men and 18 women with a mean ± SD age of 52.96 ± 12.51 years and a mean ± SD disease duration of 50.05 ± 84.9 months were included. The control group consisted of 20 healthy volunteers matched for age, sex, and other anthropometric characteristics.

In order to avoid confusion with known risk factors for atherosclerosis, the exclusion criteria comprised hypertension (systolic blood pressure [SBP]/diastolic blood pressure [DBP] of ≥140/90 mm Hg or the use of antihypertensive medication), hyperlipidemia (total cholesterol levels of ≥200 mg/dl, low-density lipoprotein cholesterol levels of ≥115 mg/dl, or triglyceride levels of ≥150 mg/dl, or the use of lipid-lowering medication), diabetes mellitus diagnosed on the basis of the World Health Organization criteria (20) or the use of antidiabetic medications, and a history of ischemic heart disease or cerebrovascular events. Further exclusion criteria were a technically poor acoustic window precluding satisfactory 2-dimensional (2-D) Doppler echocardiographic imaging of the left ventricle (LV) or left anterior descending (LAD) coronary artery flow (for CFR assessment); congenital, valvular, or hypertrophic cardiomyopathy, myocarditis or pericarditis, or thyroid diseases; severe mental retardation; and lymphoproliferative disorders. None of the subjects were past or present smokers. There was no significant family history of CVD or glucocorticoid exposure among the subjects.

The study protocol was approved by our local ethics committee and informed consent was obtained from all of the subjects prior to the performance of any procedures. This study was conducted in accordance with the ethical guidelines of the 1975 Declaration of Helsinki.

CV evaluation

The CV risk profile of all the patients was assessed by standard electrocardiography (EKG) and conventional and stress transthoracic echocardiography with CFR measurement and carotid ultrasound evaluation. The transthoracic Doppler–derived CFR, common CIMT, and arterial stiffness parameters (PWV and the stiffness index [β]) were analyzed by 2 independent echocardiographers not involved in the care of the patients (LG and LT). Arterial blood pressure and EKG were evaluated using standard procedures (21).

Standard echocardiography

The transthoracic echocardiographic images were recorded using a commercially available ultrasound unit (iE33, Philips Medical Systems) equipped with a 1–2 MHz (S5) transducer and a 3–8 MHz broadband high-frequency transthoracic transducer (S8) with second harmonics. LV diameter and wall thickness were measured using the targeted 2-D M-mode echocardiographic trace as recommended by the American Society of Echocardiography (22). The LV was divided into 16 segments, and segmental wall motion was graded as 1 = normal, 2 = hypokinetic, 3 = akinetic, or 4 = dyskinetic; the wall motion score index was obtained by dividing the sum of the segment scores by the number of visualized segments (23). LV mass was calculated using Devereux's formula (24), and the Doppler indices of LV diastolic function were measured using standard techniques (25).

Dipyridamole echocardiography and CFR

All of the subjects were asked to avoid food and drinks containing xanthine for ≥24 hours before their transthoracic Doppler–derived CFR evaluation. While the subjects were in a stable 90° left lateral recumbent position, LAD coronary artery flow was evaluated before and during dipyridamole infusion (0.84 mg/kg over 6 minutes) using a modified 2-chamber view to identify the distal LAD artery. Coronary blood flow in the mid-distal portion of the LAD artery was measured under the guidance of color Doppler flow mapping and was EKG synchronized. CFR was calculated offline as the ratio of the peak diastolic velocity during hyperemia to the baseline diastolic velocity (for each parameter, the results of 3 cardiac cycle measurements were averaged). A CFR value of <2.5 was considered a sign of impaired coronary function. At the same time, segmental LV wall motion, EKG, and symptom arousal were also evaluated, and the LV wall motion score index was calculated during the stress test. At the end of the protocol, all of the patients received 125–250 mg of aminophylline to counteract the effect of dipyridamole. All of these recordings were digitally stored to simplify the reviewing and measuring processes.

Carotid artery ultrasonography and stiffness assessment

Carotid artery ultrasonography was performed using a MyLab 60 ultrasound system (Esaote) with a 2–9 MHz LA532E linear array transducer equipped with RF-QIMT and RF-QAS software that complies with the Mannheim consensus (26). The patients were scanned in a supine position with the neck extended and the chin turned contralaterally to the examined side in order to evaluate the carotid arteries both transversely and longitudinally. The IMT of the common carotid artery (CCA) was measured 1 cm distal to the carotid bifurcation in the posterior wall of both the right and left carotid arteries. IMT was defined as the distance between the leading edges of the lumen interfaces and the media–adventitia interface of the far wall (27, 28), and the mean was taken from 3 measurements. In order to assess stiffness, right and left CCAs were examined ∼1 cm proximal to the bulb region. Time-related pressure waveforms were obtained from the systolic and diastolic changes in arterial diameter after calibrating blood pressure measured in the right upper arm using a cuff manometer. The stiffness index (β) was automatically calculated as the mean of 5 beats using the established formula β = ln(SBP/DBP)/(Sd − Dd/Dd), where Sd and Dd are the maximum and minimum CCA diameters, respectively (29); PWV was obtained from the online 1-point measurement calculated on the basis of the time delay between 2 adjacent distension waveforms using the formula PWV = distance/time (i.e., the distance between the 2 sites at which the wave was recorded divided by the time taken for the wave to travel from the first site to the second site) (30).

Laboratory analyses

The SSc-related laboratory variables (erythrocyte sedimentation rate, white blood cell and platelet counts, and C-reactive protein levels) were measured using routine methods. Antinuclear autoantibody levels were determined by indirect immunofluorescence with HEp-2 cells. Antibodies to extractable nuclear antigens, including SSA, SSB, Sm, RNP, and Scl-70 autoantibodies, and anticentromere antibodies were detected by enzyme-linked immunosorbent assays.

Total serum cholesterol, triglyceride, and high-density lipoprotein cholesterol levels were determined using an autoanalyzer, and low-density lipoprotein cholesterol was calculated by the Friedewald formula (31). Other standard clinical laboratory tests were performed under fasting conditions on the same day as the main evaluations. Mean glomerular filtration rate was calculated using the Cockcroft-Gault formula.


It is known that chronic inflammation (32) and endothelial dysfunction, which is characterized by a reduction in the bioavailability of NO as a result of an NO synthase blockade that seems to be the first cause of the atherosclerotic process, are important factors in the development of atherosclerotic plaque (33). ADMA is the main endogenous inhibitor of all 3 NO synthases, which suggests that ADMA is involved in endothelial dysfunction and atherosclerosis (34). Plasma ADMA levels were determined using high-performance liquid chromatography as described by Teerlink (35), with minor modifications.

Linearity was assessed in the ADMA range of 0.1–20 μM. The mean correlation coefficient was >0.99. The ADMA limit of quantitation was 0.01 μM. Analytical recovery was 98%, and the interassay coefficient of variation was >3%.

Statistical analysis

The continuous variables were expressed as mean values and SDs and the noncontinuous variables as median values and interquartile ranges. The Mann-Whitney nonparametric test was used to compare the continuous variables. The data were analyzed using SAS, version 9.2. All tests were 2-tailed, and P values less than 0.05 were considered statistically significant.


Table 1 shows the characteristics of the SSc patients and healthy controls. All of the patients had diffuse SSc, but none showed lung involvement or any history or signs of either coronary artery disease or other cardiac diseases. The majority of patients were being treated with azathioprine at a mean dosage of 150 mg/day (range 50–200), 2 were taking methotrexate, and 1 was taking mycophenolate mofetil. Seven patients were also taking low-dosage nifedipine (30 mg/day) because of Raynaud's phenomenon. None of the patients had pulmonary hypertension or any other complication. All of the patients were antinuclear autoantibody and Scl-70 positive, and all had significantly higher C-reactive protein and erythrocyte sedimentation rate values than the healthy controls (P < 0.01 for both); however, there were no significant differences in heart rate, arterial blood pressure, age, or body mass index (Table 1).

Table 1. Characteristics of the SSc patients and healthy controls*
 SSc patients (n = 20)Controls (n = 20)P
  • *

    Values are the mean ± SD unless otherwise indicated. SSc = systemic sclerosis; BMI = body mass index; NS = not significant; CRP = C-reactive protein; ESR = erythrocyte sedimentation rate.

  • All values of blood pressure were recorded at the beginning of stress echocardiography.

  • Disease duration was defined as the time since the first non–Raynaud's phenomenon symptom or other.

Women, no. (%)18 (90)14 (70)
Age, years52.96 ± 12.5155.40 ± 12.93
BMI, kg/m223.86 ± 5.0223.69 ± 1.12NS
Systolic blood pressure, mm Hg132.25 ± 22.51128.75 ± 12.07NS
Diastolic blood pressure, mm Hg82.00 ± 10.8179.46 ± 8.75NS
Heart rate, beats/minute74.05 ± 10.7766 ± 11.70NS
CRP level, mg/dl7.32 ± 1.380.22 ± 0.07< 0.0001
ESR, mm/hour18.23 ± 2.846.8 ± 0.38< 0.001
Disease duration, months50.05 ± 8.49

At baseline, standardized 2-D echocardiography showed no significant alterations in either group (Table 2). Carotid plaques were detected in 2 patients (10%), but none were found in the healthy controls. All of the arterial wall measurements were significantly different between the 2 groups, and both right and left CIMT, PWV, and stiffness index (β) were significantly higher in the SSc patients (Table 3).

Table 2. Echocardiographic parameters of the SSc patients and healthy controls*
 SSc patients (n = 20)Controls (n = 20)P
  • *

    Values are the mean ± SD unless otherwise indicated. SSc = systemic sclerosis; EDV = end-diastolic volume; NS = not significant; ESV = end-systolic volume; LVEF = left ventricular ejection fraction; LA = left atrial; Mit E = peak velocity of early diastolic mitral flow; Mit A = peak velocity of late diastolic mitral flow; MV-DT = mitral valve deceleration time; IVRT = isovolumic relaxation time; sPAP = systolic pulmonary artery pressure; Sm = peak systolic myocardial velocity derived from pulse-wave Doppler tissue; Em = peak early diastolic myocardial velocity derived from pulse-wave Doppler tissue; Am = peak late myocardial velocity derived from pulse-wave Doppler tissue; TEI index (MPI) = myocardial performance index.

EDV, ml75.15 ± 18.0880.14 ± 5.11NS
ESV, ml27.35 ± 8.3328.94 ± 4.70NS
LVEF, %64.57 ± 7.6463.46 ± 4.38NS
LA, mm33.68 ± 4.6735.77 ± 4.43NS
Mit E, meters/second92.06 ± 19.5772.51 ± 9.79< 0.01
Mit A, meters/second84.87 ± 26.6958.17 ± 9.74< 0.01
Mit E/A1.20 ± 0.511.27 ± 0.24NS
MV-DT, msec202.33 ± 69.95214.26 ± 11.52NS
IVRT, msec92.76 ± 15.2786.57 ± 6.16NS
sPAP, mm Hg29.87 ± 6.1825.47 ± 7.71NS
Sm, meters/second9.30 ± 1.659.00 ± 2.17NS
Em, meters/second11.45 ± 3.9913.93 ± 2.69< 0.01
Am, meters/second10.54 ± 3.2410.73 ± 2.81NS
E/Em9.54 ± 4.475.41 ± 1.25< 0.0001
TEI index (MPI)0.49 ± 0.090.48 ± 0.19NS
Table 3. Cardiovascular parameters in the SSc patients and healthy controls*
VariablePatients, median (IQR)Controls, median (IQR)P
  • *

    SSc = systemic sclerosis; IQR = interquartile range; ADMA = asymmetric dimethylarginine; CFR = coronary flow reserve; IMT = intima-media thickness; PWV = pulse wave velocity.

  • Mann-Whitney nonparametric test.

ADMA, μmoles/liter0.86 (0.80–0.90)0.57 (0.52–0.59)< 0.0001
CFR2.69 (2.54–2.98)3.2 (3.0–3.6)0.0033
Right IMT, cm0.69 (0.66–0.79)0.55 (0.49–0.61)0.0006
Left IMT, cm0.71 (0.69–0.75)0.46 (0.40–0.56)0.0006
Right PWV, meters/second7.8 (6.7–10.8)6.3 (6.1–6.9)0.0390
Left PWV, meters/second9.3 (8.0–11.1)6.6 (5.3–6.9)0.0039
Right stiffness index, β10.4 (7.6–16.3)0.95 (0.72–1.1)0.0004
Left stiffness index, β13.0 (11.5–17.2)0.89 (0.72–1.3)0.0004

The SSc patients also had significantly lower CFR (P = 0.0033) and significantly higher plasma ADMA levels (P < 0.0001) than the healthy controls (Table 3). There was no significant correlation between plasma ADMA levels and CFR. However, the dispersion plot of ADMA levels versus CFR indicated a cutoff value of 70 μmoles/liter for ADMA levels between SSc patients (above 70 μmoles/liter) and controls (below 70 μmoles/liter) (Figure 1).

Figure 1.

Dispersion plot of asymmetric dimethylarginine (ADMA) levels versus coronary flow reserve (CFR). The solid line represents the linear interpolation, and the dotted lines represent the 95% confidence limits.


Early autopsy and angiography studies found that SSc patients with symptoms of angina or acute myocardial infarction may have microvascular disease rather than coronary atherosclerosis (36, 37), but a recent study has demonstrated that atherosclerosis is not as rare as previously thought (38). In comparison to controls, our SSc patients who had no signs of or risk factors for CVD showed a significantly impaired CFR and increased carotid artery stiffness and CIMT. These results are in line with recent data concerning CV involvement in SSc. In particular, a meta-analysis of CIMT in SADs showed that CIMT was significantly increased in an SSc population compared with healthy age- and sex-matched controls (38). Bartoli et al (39) demonstrated increased CIMT and an association between higher CIMT and the deletion polymorphism of the angiotensin-converting enzyme gene in patients with SSc.

Arterial stiffness has also been examined in SSc, but with varying results (11, 40–42). Cypiene et al (40) found that arterial stiffness, as measured by pulse wave analysis and PWV, was elevated in 17 patients with diffuse SSc compared with 34 healthy controls. Timar et al (11) found that augmentation index and PWV were both elevated in 40 patients with SSc and 35 controls. However, several other studies have been unable to demonstrate a significant difference in arterial stiffness between patients with SSc and controls (41, 42). Several studies have found an impaired CFR in patients with rheumatic disease who had no evidence of CVD (43, 44); very often, the CFR was reduced but still remained above pathologic values, therefore confirming that the prevalence of microvascular involvement is high and more common than large-vessel disease (45).

Patients with diffuse SSc have subclinical CV involvement, as has been demonstrated by various invasive (5) and noninvasive methods such as ultrasonography (i.e., carotid ultrasonography and/or transthoracic echocardiography) (40, 43). Among the noninvasive techniques, echocardiography with CFR assessment plays the most important role, as has been reported in a recent review undertaken by the University of California (46). However, endothelial dysfunction has been assessed using flow-mediated dilatation (47, 48). Another study showed that flow-mediated dilatation was decreased in patients with SSc compared with controls (8). Some studies found nitrate-mediated dilatation was reduced in SSc, suggesting coexisting functional or structural abnormalities of arterial smooth muscle, adventitia, or both (48, 49). Our patient cohort had an impaired CFR with increased ADMA levels. It has recently been demonstrated that CFR is a highly sensitive (>90%) diagnostic marker of coronary artery disease, and that a CFR of <2 accurately predicts the presence of severe coronary stenosis (i.e., >70% coronary narrowing) (50). It has also been suggested that significantly increased ADMA levels may indicate endothelial dysfunction and impaired coronary microcirculation, as we have previously demonstrated in patients with rheumatoid arthritis and psoriatic arthritis (6, 51). ADMA is a marker of atherosclerosis and may predict CV risk in diseases associated with enhanced atherosclerosis, such as diabetes mellitus.

The data from our patients without conventional risk factors for atherosclerosis led us to conclude that ADMA is a newly defined risk factor in SSc and other SADs. High ADMA levels in SADs may be due to a number of mechanisms, such as increased endothelial cell turnover and the ensuing liberation of free ADMA during protein catabolism; this result offers the hypothesis that various mechanisms are involved in SSc. These findings might support the suggestion that SSc may be associated with an increased CV risk similar to that of other SADs. Moreover, while CIMT and stiffness are relatively stable, vasodilatation can be affected by many confounding factors, such as disease duration and patient age. Furthermore, the majority of our patients were taking immunosuppressive drugs, which have an antiinflammatory effect that potentially affected the atherosclerosis process and, therefore, our results (52, 53). The main limitation of this study was the small number of patients; this, in itself, may be due to the low prevalence of SSc (between 3 and 24 cases per 100,000 subjects) (45, 46) and the size of our cohort that, however, was similar to the size of other studies.

Patients with diffuse SSc seem to have subclinical CV involvement, as demonstrated by various methods. Further studies are required to define more precise algorithms for assessing and managing CVD in SSc patients. Multispecialist collaboration (i.e., between immunologists, rheumatologists, and cardiologists) and support from national and international health agencies are necessary to prevent CV complications in diffuse SSc and other SADs.


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. Turiel 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. Turiel, Gianturco, Sarzi-Puttini, Atzeni.

Acquisition of data. Tomasoni, Epis.

Analysis and interpretation of data. Ricci, De Gennaro Colonna, Ferrario.