Sjögren’s Syndrome (SS) is one of the most frequent rheumatic diseases, with a prevalence of 0.1-0.6% in the adult population, a male-female ratio of approximately 1:9 and an average age at diagnosis of 50 years (1). It is characterized by a lymphocytic infiltrate in the lacrimal and salivary glands causing keratoconjunctivitis sicca and xerostomia, which are the main clinical manifestations of the disease (2). Other clinical manifestations are vaginal dryness, non-productive cough, salivary gland swelling and systemic symptoms (arthralgia, fatigue and general discomfort) (3, 4).

SS can be primary (pSS) or associated to other connective tissue disorders, in particular Systemic Lupus Erythematosus (SLE) and rheumatoid arthritis (RA). Extra glandular involvement may be present in about one third of patients with pSS. The most affected organs are thyroid, lungs, gastrointestinal tract, blood, kidneys, skin and central and peripheral nervous system. The involvement of the cardio-vascular system is less commonly observed in these patients. Severe complications as acute pericarditis and myocarditis have been rarely reported (5), although the best-known cardiac complication is congenital heart block associated with positive anti-Ro/SSA antibodies, which is one of the manifestations of neonatal lupus. However, the latter is rare in adults and is not consistently correlated with the presence of anti-SSA/Ro antibodies (6). Moreover, patients with SS secondary to anti-phospholipid antibody syndrome (APS) are characterized by an increased risk of ischemic stroke.

Numerous studies have highlighted how acute systemic inflammation and chronic vasculitis are known to be associated with endothelial dysfunction (7). Furthermore, in some rheumatic diseases including SLE and RA, the tendency to develop atherosclerosis is one of the major causes of mortality (8).

Primary SS appears to be associated with a greater “overall risk” of CV and cerebrovascular events. However, evidence related to prevalence of cerebrovascular events appears conflicting. In fact, pSS can be associated with vasculitis, which may cause an increased risk of haemorrhagic strokes due to the inflammatory infiltrate in the vascular wall (9). However, other studies have reported either a greater risk of ischemic cerebrovascular events (10) or have shown no difference between patients with pSS and the control group (11).

The purpose of this review was to evaluate the risk of CV risk in pSS and the effect of medications on this manifestation.

Although not conventionally considered a feature of the disease, CV events represent a major burden in pSS patients, similarly to what has been extensively described in other autoimmune diseases such as RA and SLE.

Several studies demonstrated an increased risk of CV and cerebrovascular events in pSS patients, compared to the general population. The main limitation to the interpretation of these data is the wide heterogeneity of outcomes and of variable definitions. In fact, the definition of cardio and cerebro-vascular events, beyond stroke and acute myocardial infarction (AMI), may include or exclude transient ischemic attacks (TIA), arrhythmias, angina, valvular disease and thromboembolic events.

However, the prevalence of CV involvement in a very wide sense was demonstrated to be about 61.6% in pSS subjects compared to 29.7% in healthy controls (HC) (12). More specifically, in a cohort study, cerebrovascular and CV events had a significantly higher prevalence in pSS compared to HC (2.5% vs 1.4% and 1.0% vs 0.4%, respectively) (13). The hazard ratio of pulmonary embolism (PE), deep vein thrombosis (DVT) and venous thromboembolism (VTE) was demonstrated to be 4.07, 2.80 and 2.92, respectively, compared to healthy subjects, especially in the first years after diagnosis (14). These data have not been confirmed in all studies published on the topic (15), which may be due to differences in terms of ethnicity and selection of study participants.

Nonetheless, a recent meta-analysis including 9 studies, confirmed a higher risk of CV (OR=1.30, 1.09 to 1.55) but not cerebrovascular events (OR=1.31, 0.96 to 1.79). Similarly, a more recent meta-analysis including 14 studies reproduced similar results and found a higher rate of coronary disease (RR 1.34, 1.06 to 1.38), cerebrovascular disease (RR 1.46, 1.43 to 1.49), heart failure (RR 2.54, 1.30 to 4.97) and thromboembolism (RR 1.78, 1.41 to 2.25) in pSS patients. However, the higher rate of CV events was not associated with a significant increased risk of mortality (16).

The most likely reason underlying the consistency but not universal agreement of the results regarding the higher frequency of major CV events in pSS is likely mostly attributable to their time-dependency. In fact, risk factors usually require a long exposure time before actually causing the outcome. As a consequence, a very long observation time is required to detect the impact of one single risk factor, as the disease itself, especially when multiple other and more frequent factors, as other CV risk factors, affect the outcome.

In order to analyse the effect of risk factors earlier enough, surrogate outcome measures are commonly employed. A surrogate outcome measure is a measurable variable that is more sensitive to change than major outcomes but also directly associated to them. When these conditions are satisfied, a surrogate marker can partially replace the measure of the main outcome.

In the field of CV research, multiple surrogate outcome measures have been validated and demonstrated to be associated to pSS. Similarly to RA, pSS patients show a higher carotid intima-media thickness (a surrogate marker of atherosclerosis) than HC, and having pSS is an independent risk factor for arterial wall thickening (17). Another surrogate of accelerated atherosclerosis is arterial stiffness, measured as pulse-wave velocity, which has been found to be higher in pSS compared to HC (18–21). Other studies also demonstrated a lower ankle-brachial index (ABI) in pSS patients (22) and a subclinical involvement of the heart. In fact, higher prevalence of valvular regurgitation, pericardial effusion, pulmonary hypertension, increased left ventricular mass, systolic dysfunction and lower coronary reserve were demonstrated in pSS patients (18, 23, 24). Some Authors suggest that the higher prevalence of heart failure may be a consequence of ventriculo-arterial coupling, i.e. the premature return of the pulse wave caused by reduced arterial elastance may contribute to left ventricular hypertrophy and, in turn, diastolic dysfunction (24).

Although there is no doubt that pSS patients are burdened by a higher CV risk, the causes and factors accounting for it are yet to be fully understood. The most supported theory is that CV risk is the consequence of a complex combination of multiple factors, including traditional risk factors and disease-related mechanisms. The full comprehension of the single determinants is mandatory in order to be able to properly assess and address them from a clinical point of view, thus offering the best care to patients.

The first obvious consideration to make is whether pSS patients have a different prevalence of traditional CV risk factors compared to the general population, which may account for the excess CV morbidity. Indeed, multiple observational studies demonstrated a higher prevalence of hypertension in pSS (28-50% vs 11.3-25.6% in the general population) (13, 25, 26). However, other studies show opposite results (27), mostly due to differences in methodological approaches and definition of hypertension. Interestingly, one study found that hypertension tends to be under-treated in pSS patients compared to the general population (25).

Lipid metabolism seems also to be altered in pSS patients. Several studies found a higher prevalence of dyslipidaemia in pSS patients as compared to the general population (21-22.5% vs 4.2-9.5%) (13, 25, 26, 28, 29). However, findings are very heterogeneous when a more detailed analysis is carried out. In fact, some studies found higher levels of triglycerides, others lower high-density lipoprotein (HDL) cholesterol levels but not total cholesterol, others a higher rate of statin use as an indirect measure of dyslipidaemia. Additionally, a significant minority of reports showed apparently contrasting results (28). To further support the hypothesis of an altered lipid metabolism, higher levels of the adipokines – cytokines secreted by adipose tissue – resistin and adiponectin have been reported (26).

Data on metabolic syndrome, obesity and diabetes are even more conflicting and scarce. The prevalence of metabolic syndrome and diabetes seems to be higher in pSS (39.4% vs 16.9% and 27% vs 13%, respectively) (26, 27), which is in contrast with other studies reporting a lower prevalence of obesity (11% vs 21%) and diabetes mellitus (4% vs 7%) (13). As previously mentioned, these differences can globally be explained by different methodology in terms of patients enrolment and variable definitions. In contrast, it is now well established that smoking is significantly less common among pSS patients compared to the general population but also to patients with other autoimmune diseases, probably due to the worsening of sicca symptoms induced by cigarette smoke (30, 31). However, the paucity of data on this topic significantly hamper a full understanding of the picture.

Albeit SS patients seem to have an overall higher prevalence of traditional CV risk factors, this is not sufficient to account for the whole excess CV morbidity. Few studies investigated CV events in distinct subset of SS patients, thus data so far are limited and mostly derive from indirect evidence. However, some reports suggest that the role of anti-SSA/Ro and anti-SSB/La is significant to the point that the rate of cerebral infarction and VTE may be higher compared to the general population (HR = 1.7, 1.0 to 2.9 and 3.1, 1.9 to 4.8, respectively) only in the subgroup of seropositive subjects (32). Although these data need to be confirmed, there is little doubt that the presence of autoantibodies has an impact of CV morbidity. In fact, through a neural network analysis of a large cohort of SS patients, the presence of anti-SSA/Ro antibodies turned out to represent a hub, connecting the two main pillars of CV disease phenotypes in SS, i.e. ischemic (such as myocardial o cerebral infarction) and non-ischemic events (such as heart failure) (33, 34). To further support a close link between disease features and CV events, the same study showed that ischemic events were mostly associated to variables characterizing extra-glandular involvement (vasculitis, low complement, leukopenia, cryoglobulins) while non-ischemic events were mainly linked to the presence of sicca syndrome (34). Additionally, a higher prevalence of anti-phospholipid antibodies was found in SS patients, potentially contributing to CV disease (4).

In a complex disease such as SS, characterized by a wide spectrum of alterations of immunity and inflammatory processes, it is intuitive to think that a strict and compartmentalized model of CV risk into traditional and disease-related risk factors cannot reliably represent reality.

Consequently, it is important to investigate the complex relationships between disease-related features, endothelial dysfunction and traditional risk factor. In line with the findings described above, a reduction of ABI is associated to seropositivity for anti-Ro/SSA and anti-La/SSB (22) and a higher prevalence of CV events is associated to central nervous system (CNS) involvement and immunosuppressive treatment (a surrogate marker of severe disease) (13). As endothelial dysfunction is an early marker of atherosclerosis - an inflammatory process involving vessel walls - an association between metabolic syndrome, dyslipidaemia and serum concentration of interleukin (IL)-1β and IL-6 in SS patients is not surprising (26). Additionally, endothelial dysfunction seems more pronounced in patients with joint involvement, parotid enlargement (35) and, more generally, with active disease measured by European League Against Rheumatism (EULAR) Sjogren’s syndrome disease activity index (ESSDAI) (36, 37).

However, contrasting data have been published too, finding increased aortic stiffness in patients with no circulating anti-Ro/SSA antibodies (21) and an association between traditional risk factors and the presence of extra-glandular disease features (27). Other studies, instead, were not able to find any significant associations between endothelial dysfunction and disease features (38). Indeed, endothelial dysfunction is one of the earliest changes that characterize atherosclerosis. Its main features are reduced vasodilatory response, alterations in the expression of surface adhesion molecules and deficient repair mechanisms and endothelial cell turnover. All these features lead to accelerated atherosclerosis. Multiple studies have investigated endothelial changes in SS. In terms of functional assays, SS patients display a deficient flow-mediated dilation (FMD) (38, 39) and nitrate-mediated vasodilation (NMD) (35). These functional aspects are in agreement with higher levels of circulating asymmetric dimethylarginine (ADMA), responsible for reduced production of nitric oxide in SS compared to HC (18, 39).

As far as endothelial damage repair mechanisms are concerned, SS patients have an increased number of circulating endothelial microparticles (EMP), endothelial progenitor cells (EPC) (40) and angiostatin (39). Very interestingly, EPCs, involved in endothelial repair, seem to be higher in early disease, while EMPs, a marker of damage, in late disease, which leads to the hypothesis that repair mechanisms remain functional in early disease and wane at later stages, potentially contributing to accelerated atherosclerosis (40). Additionally, a recently described T cell population involved in the coordination of endothelium repair mechanisms, thus named angiogenic T cell (Tang), is expanded in peripheral blood and salivary glands of SS patients and may play a role in the induction or persistence of endothelial dysfunction (36).

It is therefore clear that to fully understand the determinants of CV risk in SS, we cannot disregard the intricate connections among the multiple players involved. Due to the unavoidable influence they exert on each other, dissecting the role of each factor independently is unreasonable. Thus, a multifaceted approach to the topic is the only realistic path to further extend our knowledge in this complex field, ultimately to improve patient care.

Atherosclerosis is a multifactorial disorder that tends to begin early in life, but with late-onset clinical manifestations. The isolation of immune cells including lymphocytes and macrophages from atherosclerotic lesions indicates an involvement of the immune system in the etiopathogenesis of this disorder (41, 42) ( Figure 1 ). Inflammation may play a crucial role in the exacerbation of atherosclerosis. Systemic autoimmune rheumatic diseases (AIRDs) lead to an increased risk of morbidity and mortality due to cardiovascular events, especially following atherosclerotic events. This could be attributed to the traditional risk factors for atherosclerosis, and to the treatment with specific drugs such as corticosteroids (43, 44).

The infiltration of atherosclerotic plaques by immune cells, as well as their activation, can occur as a result of numerous trigger factors, such as microbial infections (45). It is thought that these cells have a fundamental role in promoting the progression of atherosclerosis and this is based on several studies. For instance, in C57BL/6 mice the depletion of CD4 + and CD8 + T lymphocytes may lead to a net decrease in the formation of lipid striae. Furthermore, mice with SCID (severe combined immunodeficiency) and apolipoprotein E (ApoE) knockout mice, showed a reduction in aortic lipid striae in 73% of cases, compared to mice with a normal immune system. Finally, the transfer of CD4 + T lymphocytes to immunodeficient mice led to an increase in the dimensions of atherosclerotic lesions (46). Therefore, it is clear that atherosclerosis has an autoimmune setting, with immunity cells releasing numerous cytokines within the plaques [interleukin (IL), tumour necrosis factor (TNF)-alpha, and Platelet-Derived Growth Factor (PDGF)].

During the atherosclerotic process, immune cells may respond specifically to certain molecules, including oxidized low-density lipoproteins (OxLDLs), heat-shock proteins (HSP), and Beta2 glycoprotein-I (β2GPI) (41). This is shown by an increase of these molecules in the atherosclerotic plaques (47). On the other hand, some autoantibodies have been associated with a greater risk of atherosclerosis, as evidenced in some animal studies carried out with anti-cardiolipin antibodies (48). OxLDLs, once phagocytised by macrophages, cause the transformation of macrophages into foam cells, in turn promoting the progression of atherosclerosis. Anti-OxLDL antibodies are present in both healthy patients and patients with AIRDs (49), but in the latter, especially in patients with SLE, SSc and systemic vasculitis, these antibodies appear to be present at higher levels than controls (49–51). A correlation between anti-OxLDL antibody levels and total immunoglobulin levels was also observed in SLE patients (50).

The immune system dysregulation plays an important role in the pathogenesis of early atherosclerosis in patients with pSS. The positivity of anti-SSA/Ro and -SSB/La antibodies, two well-known markers of the disease and generally associated with systemic manifestations (35, 52), may correlate with endothelial dysfunction and with an increase in the intima-media thickness. The latter, along with alteration in the NO-induced vasodilatation, may be associated with and predictive for leukopenia, marker of high disease activity and higher risk of CV events (13, 35, 52). The organic and functional damage of the smooth muscle cells in the vessel tunica media may be determined by cellular infiltration of the sub-endothelial space. This is demonstrated by the higher levels of ICAM-1 and VCAM-1 adhesion molecules found in patients with SS. These molecules may recruit leukocytes with consequent infiltration of the vascular wall and induction of atherosclerotic damage (35).

The real contribution of inflammation in the pathogenesis of atherosclerotic damage in SS is not fully understood. Although two studies showed a correlation between C-reactive protein (CRP) levels and the risk of developing atherosclerosis and CV events (18, 25, 27), most studies found no differences in CRP levels in patients with SS compared to HC (17, 27, 35, 38, 53–55). An important role is played by calprotectin (53), a protein expressed in the cytoplasm of neutrophils and an excellent indicator of innate immune activity. This protein is highly expressed in patients with SS in both glandular sites, correlating with the Focus Score (FS) and the circulation, being associated with the increased production of cytokines such as IL-1β, IL-6, TNF-α, IFN-γ, IL-10 and IL-22. This molecule seems to represent an important prognostic factor for CV events, regardless of other risk factors (56). Another noteworthy mechanism is the Wingless-type (Wnt) signaling pathway, which plays a key role in the process of inflammation regulation (17). In fact, there is a correlation between the DKK-1 molecule (Dickkopf WNT signalling pathway inhibitor 1) and atherosclerosis. This molecule appears to be more expressed in endothelial cells and atherosclerotic plaques, where it promotes the interaction between platelets and the endothelium, in turn causing local inflammation and plaque destabilization (57). Recent studies on the (P2X7R)-NLRP3 complex involved in the pathogenesis of sialadenitis in SS showed a close correlation between inflammation and infiltration of the glands and those observed in the vessel wall in atherosclerosis (58).

Therapeutic strategies for atherosclerosis and for any other CV disease should be evaluated on the basis of the specific clinical manifestations (59). In patients at increased risk of thrombosis and vascular occlusion, such as in SLE and APS, antiaggregating therapy (aspirin) should be recommended (60). The key point of the treatment is prevention (61). The increased risk of developing atherosclerosis or other CV events in patients suffering from autoimmune diseases is caused not only by the autoimmune state itself, but also by the complications of these diseases and by the treatments carried out to treat them (for example corticosteroids) (62). Careful monitoring of blood pressure, periodic checks of laboratory parameters and lipids, regular exercise and the use of drugs, such as statins and folic acid, if necessary, are some of the main therapeutic strategies to prevent the formation or evolution of these pathological conditions (63). Further approaches are still under study, such as immunomodulation (immunosuppressive drugs, administration of IV immunoglobulins (IVIg), induction of tolerance with autoantigens) (64), use of inhibitory cytokines, gene therapy and bone marrow transplantation.

The management of pSS has not undergone significant changes in the last few decades and continues to rely on the symptomatic treatment of sicca syndrome and on the use of a variety of immunosuppressive drugs to treat its systemic manifestations (65). Although the great amount of drugs used to treat the systemic manifestations of the disease, data on CV manifestations are scarce and extrapolated according to their mechanisms of action, as summarized in Table 1 .

Several studies focused on conventional systemic immunosuppressive therapy (prednisone, cyclosporin, azathioprine, methotrexate) and other drugs (dehydroepiandrosterone, nizatidine, rebamipide), have not demonstrated efficacy in the control of syndrome sicca, although mild benefits have been reported for some drugs (1, 66, 88–90). Furthermore, conventional and biological DMARDs, widely used in other rheumatological diseases, did not show significant clinical effects in pSS. Therefore, their use remains small and limited to some specific patient subsets (91).

Some clinical studies have been carried out using TNF-α (infliximab and etanercept) and interleukin 1 (IL-1) (anakinra) as targets, but with poor results (124–126). A recent study evaluated the efficacy of tocilizumab (anti-IL-6) in patients with SS, but it did not show an improvement neither in sicca symptoms nor in the systemic manifestations of the disease compared to placebo (127). Similarly, abatacept, a drug acting as inhibitor of CD80/CD86-CD28 co-stimulation, has not shown positive results, despite some previous studies reported an improvement in disease activity (assessed by ESSDAI) and a reduction in cytokine and autoantibody levels (128, 129).

In conclusion, surely pSS patients should receive a complete diagnosis, aimed at determining the extent and severity of the disease, the subset of the disease and the impact of the fatigue. The therapeutic approach is substantially similar in case of pSS or associated SS. Recent published recommendations include non-pharmacological and preventive interventions above all, such as hygiene measures, abstention from cigarette smoking, and counseling on a correct diet and appropriate use of drugs, etc. In addition, artificial tears and artificial saliva can be used, and if symptoms do not resolve, some patients may benefit from secretagogue drugs (pilocarpine and cimeveline) (166). In patients with moderate/severe involvement, a number of immunosuppressive drugs (synthetic and biological DMARDs) can be used. RTX should be used in case of severe organ manifestations not responsive to other treatments, such as cryoglobulinemic vasculitis. Treatment of fatigue primarily involves a low-impact aerobic exercise program. In patients with depression, anxiety or attention deficit, tests must be carried out in order to identify those who will need specific therapies. The treatment of the main cardiopulmonary manifestations (interstitial pneumonia, pulmonary hypertension and myocarditis) is based on the use of high-dose GCs often in combination with MMF or cyclophosphamide (CYC), or even RTX can be considered. Finally, the treatment of lymphoma in patients with SS should be carried out following the same protocols used for patients without SS. However, although most of the drugs used in the treatment of Sjogren’s syndrome have shown positive effects on different outcomes of the disease, to the best of our knowledge, we have poor data about the role of these drugs on CV events and surely further studies are needed to investigate the effect of these therapies on CV outcome in pSS patients.

FA, FG, GC, EB, and CP designed and performed the literature search. FA and EB drafted the narrative review and edited the manuscript. All authors read the manuscript, revised it for intellectual content, approved the final version, and agree to be accountable for all aspects of the work.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.