Vascular stiffness is usually estimated from FMD or PWV using vascular ultrasonography. A higher pulse pressure to achieve a certain degree of vessel distension suggests a lower vessel distensibility and thus a higher stiffness (121). Impaired FMD is thought to represent an early stage in the process of atherogenesis. SLE patients have significantly lower FMD, not fully explained by classic cardiovascular risk factors (111). Lupus nephritis, higher disease activity measured with SLE Disease Activity Index (SLEDAI), and higher levels of CRP have been correlated with impaired brachial endothelial function (122). Aortic stiffness, evaluated directly with transoesophageal echocardiography, is higher in SLE patients with hypertension, but normotensive SLE patients also have higher aortic stiffness than normotensive controls (123).

Systemic lupus erythematosus patients also have higher PWV (carotid-femoral or carotid-radial) than controls, but the predictive value for CVE is not clear (113). Despite that, PWV velocity has been associated with other classic cardiovascular risk factors such as age, body mass index, and hypertension (113–115, 124).

Most studies determine IMT at the carotid level due to the better accessibility, but the prevalence, characteristics, risk factors, or predictors of aortic atherosclerosis are probably similar to carotid atherosclerosis (125–127). There is considerable heterogeneity regarding studies protocols in the method of measuring IMT in terms of the carotid segment investigated, unilateral/bilateral measurements, far wall, near wall, or both walls measurements and expression and mean or a maximal IMT. Nevertheless, there is clear evidence for a significant increase of cIMT in SLE patients, as showed in a systematic review and meta-analysis (116). In women with SLE without previous CVE, the cIMT at baseline was predictive of future CVE (117). This was especially true in the presence of concomitant carotid plaques. As for aortic IMT, values are also higher in SLE patients than age- and gender-matched controls (117).

In SLE, carotid plaques are reported to be 2.4 times higher than the general population, with a peak of 5.6 times higher among patients <40 years (84). Patients at higher risk also include those with longer disease duration, higher damage index score, and less aggressive immunosuppressive therapies (84). Likewise, coronary artery calcifications evaluated through computed tomography appear to be more pronounced in younger SLE patients when compared to age-matched controls, but also in patients with increased disease activity (128). When assessing the risk of CVE related to atherosclerotic plaques, it has been suggested that quantification of plaques may reflect better the extent of atherosclerosis, rather than only assessing the presence or absence of plaques (109).

Using SPECT, myocardial perfusion defects were detected in 40% of women with SLE (118). A 13-fold increased risk of CVE was found in patients with SLE who had perfusion abnormalities on SPECT after a mean follow-up of 8.7 years (120). In other study using positron emission tomography, myocardial blood flow was measured in the resting phase and after administration of adenosine (hyperemic phase) (119). Inclusion criteria were patients suffering from either SLE or rheumatoid arthritis, with normal or minimally diseased coronary arteries (<20% luminal diameter reduction) at angiography and without cardiovascular risk factors. Overall, the study found that myocardial blood flow during adenosine hyperemia and coronary flow reserve were significantly blunted in patients compared with controls. Despite the small sample size, the study alerted for the fact that a reduced coronary flow reserve can be found in the absence of significant coronary artery disease and might be the result of prolonged inflammation that precedes premature coronary artery disease.

Myocardial perfusion studies using MRI have, additionally, led to conclude that SLE patients tend to have a distinct and more diffuse pattern of coronary artery wall contrast enhancement than coronary artery patients without SLE or healthy controls. These differences were explained by a more diffuse vascular inflammation in SLE (129).

Carotid ultrasound will remain probably the most used imaging method to assess atherosclerosis as it is relatively non-inexpensive and non-invasive and already validated. Still, there is great variation in IMT acquisition and analysis among studies. Although for SLE no specific recommendations exist for assessment of IMT, compliance with other published guidelines, such as from the American Society of Echocardiography cIMT task force (2012) or the Advisory Board for the “watching the risk symposia” at the European Stroke Conferences (2011), can help to standardize methods of assessment, improve the quality of future studies, and facilitate the comparison of results among them (130, 131). Similarly, the International Brachial Artery Reactivity Task Force guidelines should be applied when assessing FMD of the brachial artery (132). A combined approach, for example, assessment of IMT and plaque, may improve sensibility and risk prediction.

As described elsewhere in this article, traditional risk factors could not account for the total burden of CVE in SLE; however, they are still very important and frequent in this population. In the Hopkins Lupus Cohort, several traditional risk factors like age, hypertension, obesity, and cholesterol >200 mg/day were showed to be the main predictors of coronary artery disease (109). Despite being a relatively young SLE cohort (average 38.3 ± 12.1 years), half the patients had three or more coronary artery disease risk factors, largely exceeding that of other same-nationality populations matched for age, race, and sex. Other cohorts have also shown, compared to healthy controls, a higher prevalence of sedentary lifestyle and obesity (85), older age (12, 14, 133), hypertension (14), dyslipidemia (12), and DM (85).

Tobacco use has also been linked to coronary events in SLE (7, 134), although other studies failed to prove this (135). Differences among the cardiovascular risk associated with smoking have been hypothesized to be due to race specific effects, as one study showed that history of smoking in SLE was associated with higher mean IMT measures in black women, but not white woman (136).

Disability caused by SLE can limit the ability to exercise. A sedentary life style has been identified as one of the most common risk factors for coronary artery disease, being present in 70% of patients (135). This was attributed to disease and treatment-related variables such as fatigue, anemia, arthritis, and avascular necrosis of bone.

There has been a long debate on the theoretical dual action of corticosteroids in atherosclerosis. Since inflammation plays a significant role in atherogenesis, one might think that the anti-inflammatory action of corticosteroids would reduce this risk. On the other hand, corticosteroids enhance classical cardiovascular risk factors such as hypertension, hyperglycemia, dyslipidaemia, and obesity (151). Different variables such as duration of treatment, cumulative, average (all time, only few years) or maximum doses have been evaluated in different studies, which can in part explain the mixed results. A positive correlation has been found with both duration of treatment (152) and cumulative dose (10, 148, 187).

When addressing the causative role of corticosteroids, one shall also consider that cumulative disease activity and some clinical phenotypes, such as nephritis and neuropsychiatric lupus, often warrant high doses of corticosteroids and have also been associated with more atherosclerosis in SLE and thus could be confounding factors. After adjusting for all these confounding factors, Manzi and colleagues demonstrated that a longer treatment duration with prednisolone (but not current use, maximum dose, or cumulative dose) was an independent determinant of carotid plaque, while disease activity (SLAM) determined at the time of the detection of the plaque was found to be inversely related to the presence of plaque (109). Completely different results were obtained by another study that reported current disease activity (SELENA-SLEDAI) and current dose of corticosteroids (>20 mg/day) were associated with a higher cardiovascular risk (11). Finally, it has been purposed that the action of corticosteroids may be dose-dependent, as one study in pediatric SLE showed an association between the highest and lowest cumulative doses of corticosteroids and a higher cIMT, while moderate cumulative doses were associated with a decreased cIMT (188).

Statins theoretical benefits are related to their effect in decreasing proinflammatory cytokines and chemokines such as IL-6, IL-8, TNF-α, and MCP-1 (153, 154, 157, 189). In a Chinese randomized, double-blind, placebo-controlled trial, low-dose rosuvastatin (10 mg/day) was associated with reduction of LDL, CRP, P-selectin, and thrombomodulin (158). Moreover, atorvastatin (20 mg/day) therapy for 8 weeks has shown to reduce arterial stiffness of SLE female patients who had baseline pathological carotid-femoral PWV, although this was only significant for middle-aged patients (36–59 years) (159). Atorvastatin at a dose of 40 mg/day may stabilize the coronary artery calcium score, although it did not ameliorate perfusion defects in myocardium SPECT (190). Results from the 2-year double-blinded Lupus Atherosclerosis Prevention Study were, unfortunately, more disappointing (155). In this study, 200 SLE patients without previous cardiovascular disease were randomized to receive atorvastatin (40 mg/day) or placebo. Helical-CT scanning for coronary artery calcium measurement and cIMT and plaque detection were performed at baseline and after 2 years, but no significant difference was found between the groups in terms of progression of these variables. Similar negative results have also been obtained in SLE pediatric populations (156).

The renin–angiotensin–aldosterone system (RAAS) has been implicated in atherogenesis (191). In a SLE mice model, an enhance vasoconstriction response to the RAAS which promotes vascular changes during the SLE course was noticed (192). Despite this possible contribution for atherosclerosis development, only few studies have focused on the potentially protective role of ACEIs and ARBs in SLE. A non-randomized prospective trial, comprising 144 patients with lupus nephritis treated with ACEIs/ARBs and 301 non-treated patients, showed no significant differences in the cumulative occurrence of CVE between the groups (160). One cross-sectional study comprising 51 SLE patients, mainly African-American, of whom only 12 were being treated with ACEIs, showed a strong association between total plaque area and 25(OH)-vitamin D insufficiency or ACEI non-use (161).

In healthy women, aspirin had no significant effect on the risk of myocardial infarction or death from cardiovascular causes, with the exception of women 65 years of age or older (relative risk 0.66, 95% CI 0.44–0.97, p = 0.04) (193). In SLE, aspirin did not alter the level of several atherosclerosis biomarkers (homocysteine, high-sensitivity CRP, soluble VCAM-1, P-selectin, and thrombomodulin) (158). Moreover, in the SLE cohort of the SOLVABLE study, baseline aspirin was even correlated with progression of coronary artery calcifications (194). Nevertheless, the authors hypothesized that patients under aspirin were already identified as those who may benefit from aspirin use. The role of aspirin in patients with APS positivity should not be forgotten, as in these patients, aspirin reduces the cardiovascular risk and thrombotic events (164).

Hydroxychloroquine (HCQ) has well-known benefits for lipid (166) and glycemic control (166, 167) and even reduces the risk of thrombovascular events (169, 170). In an animal model of SLE, early treatment with HCQ prevented the development of endothelial dysfunction via reduction of reactive oxygen species (165). Furthermore, HCQ has been shown to reduce the carotid plaque burden and aortic stiffness in SLE patients (170). One recent paper from Fasano and colleagues supported the association of HCQ and aspirin in patients with Lupus for primary prevention of CVE (162). The authors performed an observational study and multivariate analysis that lead to the conclusion that both aspirin and HCQ reduced the risk of the first CVE (hazard ratio 0.24 and 0.027, respectively), and they found a time-dependent effect of HCQ, as HCQ protective effect was only significant after 5 years of treatment. The same author found in a similar designed study that the association of aspirin to HCQ had a synergistic thromboprotective effect (163).

In mice, MMF slows down the progression of atherosclerosis by inhibiting CD4+ T-cell activation and infiltration to the atherosclerotic lesion (173). One interesting clinical study enrolled 22 SLE patients who were undergoing carotid endarterectomy and randomized them in two groups (171). One group received 1,000 mg of MMF for 2 weeks prior to the surgery, while the other group received placebo. When compared to the placebo group, the carotid plaques of the MMF group had reduced number of activated T-cells and increased number of regulatory T-cells and had also a reduced pro-inflammatory and metalloproteinase genes expression. Contrary to this trial, no beneficial effect of MMF in the progression of cIMT or coronary calcification was noted in a 2-year longitudinal cohort study, despite the fact that only 25 patients of the study received the drug and at variable doses (172).

Several data point out that azathioprine is linked to a higher risk of cardiovascular disease (40, 138, 174, 175), but as for corticosteroids, azathioprine use is also associated with higher disease activity which may cofound the results.

Studies on other less commonly used immunosuppressive drugs are scarce. The prevalence of abnormal aortic IMT and plaques was found to be negatively correlated with cyclophosphamide therapy (123). Similarly, current use of cyclosporine A was found to be protective against increased cIMT (148).

CD20-specific monoclonal antibodies administered to apoE−/− and LDLr−/− mice were shown to significantly decrease atherosclerosis, probably through reduction of the IgG type anti-oxLDL antibodies and through reduction in the accumulation of B-cells, macrophage, and T-lymphocytes in atherosclerotic plaques (180).

Mice treated with antibodies against the B-cell activating factor receptor (BAFFR) or that lacked the BAFFR exhibited reduced atherosclerosis (176). This was postulated to be related with the consequent depletion of the atherogenic B2 cells subtype and the preservation of B1 cells subtype. This last subtype is considered atheroprotective, as it produces IgM antibodies against oxLDL and apoptotic cells (177, 178). BAFF inhibition also prevents APS in lupus-prone mice, suggesting it could have a potential benefit in preventing thrombosis in patients with SLE (179).

Vitamin D regulates several important immune functions and its deficiency has been linked to premature atherosclerosis (182–185). In apolipoprotein E knockout mice, calcitriol treatment changes the function or differentiation of dendritic cells and regulatory T-cells, decreases the production of pro-inflammatory chemokines, and reduces the quantity of inflammatory effector cells in atherosclerotic plaques (181). For patients with SLE, however, evidence supporting the supplementation with vitamin D to reduce atherosclerosis progression are lacking. In fact, several studies failed to prove a protective effect in atherosclerosis (96–98).

Smoking cessation should be recommended for all SLE patients, because smoking is a strong predictor of cardiovascular diseases with an odds ratio of 3.731 (CI 1.39–10.0) (195).

To our knowledge, only one non-randomized small-scale study analyzed the effect of supervised physical exercise on the endothelial function and ergospirometric test variables in SLE. Improved FMD, exercise tolerance, and threshold velocity after training was noted in the group of physical exercise (196).

No strong conclusion can be drawn from the currently available literature concerning the effects in atherosclerosis of most of the currently used drugs in SLE. Despite the lack of recommendations, it is prudent for clinicians to treat modifiable risk factors whenever possible.

Understanding that atherogenesis is an inflammatory process by nature is important for the future development of drugs aimed at stopping or reversing plaque formation. A meta-analysis concluded that regression/progression of cIMT induced by several drugs did not predict changes in the occurrence of major CVE (197), therefore, addressing both the reduction of atherosclerosis surrogate markers and CVE will be crucial for treatment recommendations.

Celastrol is a potential future candidate for slowing down atherosclerosis in SLE. Although not used currently for the treatment of SLE, the effect of this drug was tested in active chromatin-induced SLE BALB/c mice, where it was shown to improve proteinuria, lower anti-nuclear, and anti-dsDNA antibodies, reduce renal histological changes, and increase survival rate (198). In addition, celastrol inhibits several important atherosclerosis pathways such as the production of LDL, expression of VEGF, formation of lectin-like oxidized LDL receptor-1, and reactive oxygen species (199, 200). Plaque ratio reduction was achieved with this drug in a rabbit experimental model (199).

The 4-F peptide, an apoA-I fragment, was shown to convert HDL from pro-inflammatory to anti-inflammatory, prevent the inflammation induced by oxidized lipids and to dramatically reduce atherosclerosis in mice, suggesting that it could also be used in the future for the prevention of atherosclerosis (201, 202).

We have showed before that IL-1 is involved in atherosclerosis development, thus targeting the IL-1 pathway could be a potential therapeutic approach against atherosclerosis in patients with SLE. The Canakinumab Anti-Inflammatory Thrombosis Outcomes Study trial provided evidence that targeting IL-1β with canakinumab (150 mg every 3 months) was associated with significant lower incidence of nonfatal myocardial infarction, nonfatal stroke, or cardiovascular death (hazard ratio versus placebo, 0.85, p = 0.02) (203). The higher incidence of fatal infections noted in the canakinumab group, the lack of an established role in the treatment of SLE and their expensive cost will, however, limit the use of IL-1 blockers for preventing atherosclerosis in SLE.

If promising results are obtained in these trials, subsequent studies in SLE may be warranted.