Article Text
Abstract
Immune-mediated inflammatory diseases (IMIDs) are recognised risk factors for accelerated atherosclerotic cardiovascular disease (CVD), particularly in younger individuals and women who lack traditional CVD risk factors. Reflective of the critical role that inflammation plays in the formation, progression and rupture of atherosclerotic plaques, research into immune mechanisms of CVD has led to the identification of a range of therapeutic targets that are the subject of ongoing clinical trials. Several key inflammatory pathways implicated in the pathogenesis of atherosclerosis are targeted in people with IMIDs. However, cardiovascular risk continues to be systematically underestimated by conventional risk assessment tools in the IMID population, resulting in considerable excess CVD burden and mortality. Hence, there is a pressing need to improve methods for CVD risk-stratification among patients with IMIDs, to better guide the use of statins and other prognostic interventions. CT coronary angiography (CTCA) is the current first-line investigation for diagnosing and assessing the severity of coronary atherosclerosis in many individuals with suspected angina. Whether CTCA is also useful in the general population for reclassifying asymptomatic individuals and improving long-term prognosis remains unknown. However, in the context of IMIDs, it is conceivable that the information provided by CTCA, including state-of-the-art assessments of coronary plaque, could be an important clinical adjunct in this high-risk patient population. This narrative review discusses the current literature about the use of coronary CT for CVD risk-stratification in three of the most common IMIDs including rheumatoid arthritis, psoriasis and systemic lupus erythematosus.
- Computed Tomography Angiography
- Diagnostic Imaging
- Coronary Artery Disease
- Atherosclerosis
- Inflammation
Statistics from Altmetric.com
Introduction
People with immune-mediated inflammatory diseases (IMIDs) have a higher risk of atherosclerotic coronary artery disease (CAD) than the general population. Individuals with IMIDs have a greater burden of CAD, arising at younger age and with faster progression than those without. For systemic lupus erythematosus (SLE), this difference translates into at least a threefold increased risk of myocardial infarction (MI). Similarly, rheumatoid arthritis (RA) is associated with a twofold increased MI risk, and other IMIDs such as psoriasis also confer added risk.1 These statistics are reflective of the critical role inflammation plays in the pathogenesis of atherosclerotic cardiovascular disease (CVD), which remains the leading cause of mortality in patients with IMID. However, the excess CVD risk associated with IMIDs is not captured by current clinical risk assessment tools.
The 2023 update of National Institute for Health and Care Excellence guideline CG181 on CVD risk assessment recognises the systematic underestimation of CVD risk in people with IMIDs with standard risk assessment tools. As there is presently no established alternative, opportunities for prevention strategies are limited and this contributes to worse outcomes.2 Coronary CT is an attractive test for patients with IMID with borderline indication for statins or other preventative therapies; it is non-invasive, relatively quick to perform and rapidly becoming widely accessible. The prognostic value of coronary artery calcium (CAC) scoring is widely recognised. Moreover, the ability of CT coronary angiography (CTCA) to detect early subclinical CAD, quantify non-calcified plaque burden and identify vulnerable plaque features as well as pericoronary artery adipose tissue inflammation could further inform the cardiovascular management of patients with IMID who may lack traditional risk factors. However, data on the use of CTCA for improving CVD risk-stratification in IMIDs are sparse (figure 1 and table 1). Clear guidance to define its future role in clinical practice is needed.
This narrative review discusses the current literature relating to the use of coronary CT for CVD risk-stratification and monitoring in three IMIDs, that is, RA, psoriasis/psoriatic arthritis and SLE. RA is the most common form of inflammatory arthritis, while psoriasis is strongly associated with cardiometabolic syndrome, and SLE carries one of the highest atherosclerotic CVD risks. The role of CTCA for investigating premature CAD and coronary arteritis in some of the rarer vasculitides is outside the scope of this article.
Mechanisms of increased atherosclerotic CVD risk in IMIDs
Increased prevalence of traditional CVD risk factors
Traditional CVD risk factors including hypertension, dyslipidaemia, insulin resistance, central obesity and smoking are prevalent in IMIDs,3 but do not solely account for the excess cardiovascular risk. The role of traditional risk factors for CVD may not be directly translatable in IMIDs. In RA, lower total and low-density cholesterol are paradoxically associated with greater inflammation and CVD risk. While mechanisms of accelerated atherosclerosis in IMIDs remain incompletely understood, several key mediators associated with inflammatory disease activity and atherosclerotic CVD in IMIDs have been identified, which are briefly summarised here.
Chronic local and systemic inflammation
Immune mediators with shared mechanisms in IMIDs and atherosclerosis include interleukin (IL)-6 in RA, type I interferon (IFN-I) in SLE and IL-23/IL-17 in psoriasis. In RA, serum IL-6 is associated with vascular dysfunction and progression of subclinical atherosclerosis. IL-6 is an inflammatory cytokine with a broad range of actions, including important effects on atherogenesis. In a post-hoc analysis of CANTOS, the greatest reduction in major adverse cardiovascular events after treatment with a monoclonal antibody to IL-1β was seen in individuals who achieved lower downstream IL-6 and/or C reactive protein.4 IFN-I (mainly IFNα and IFNβ) is not only a critical driver of SLE disease activity but may also promote atherosclerotic plaque progression and destabilisation directly via its actions on plaque macrophages.3 Similarly, IL-17 is known to be upregulated in psoriasis and is also thought to have a proatherogenic role via stimulation of inflammatory macrophages and/or endothelial dysfunction. Inhibition of IL-17A has been shown to improve endothelial function in patients with psoriasis.5
Disease-specific factors
Autoantibodies are detectable in several IMIDs, including rheumatoid factor and anticyclic citrullinated peptide in RA, and antinuclear antibody and anti-double-stranded DNA in SLE. Studies that investigated these autoantibodies as potential mediators of vascular injury have yielded mixed results, with no clear association identified.
Medication effects
Antirheumatic treatments can impact the cardiovascular risk of people with IMIDs. A meta-analysis of 236 525 patients with RA showed a 47% increase in cardiovascular events among patients taking long-term steroids, suggesting that their deleterious effect on lipid metabolism, glucose tolerance and blood pressure may outweigh their anti-inflammatory function on the vascular wall. Conversely, some biological disease-modifying antirheumatic drugs (DMARDs) such as tumour necrosis factor inhibitors (TNFi) could potentially reduce the risk of cardiovascular events.6 Other biological agents used in RA such as tofacitinib, a Janus kinase inhibitor, may increase cardiovascular events.7 Further studies evaluating the effects of immunomodulatory therapies used in IMIDs on cardiovascular risk are ongoing.
Rheumatoid arthritis
RA is the most common inflammatory arthritis, affecting up to 1% of individuals aged 65–75 years old in the UK. Despite advances in the management of RA, CVD mortality remains a major challenge.2 Individuals with longer RA disease duration and time spent in flare have the highest CVD risk due to the cumulative effects of inflammatory-associated vascular injury/dysfunction.8 Data from a large registry study in Denmark showed that patients with active seropositive RA referred for CTCA to investigate chest pain had more major adverse clinical events over 3.5-year follow-up than individuals without RA.9 Histological studies have also shown that people with RA are more likely to have coronary lesions with vulnerable plaque features.
Risk-prediction scores versus coronary CT in RA
The European Alliance of Associations for Rheumatology (EULAR) recommends that people with RA undergo CVD risk estimation at least once every 5 years. Recognising that standard clinical CVD risk scores have limited use in RA, EULAR recommends adding a 1.5× multiplication factor. However, the application of a crude conversion factor across all people with RA misclassifies CVD risk in up to 10–26% when evaluated against CAC score.10 While IMID-specific risk assessment tools have been developed, including the Expanded Risk Score in Rheumatoid Arthritis, none have been shown to perform better than traditional CVD risk scores. In a study of 44 asymptomatic women with RA, neither standard nor modified systemic coronary risk evaluation (SCORE) tools could predict the finding of severe CAD on CTCA, which was observed in 9% of patients.11
Prevalence of CAC in RA
Observational studies have shown that patients with RA have a twofold to fivefold greater chance of having CAC than those without RA. In one report, up to 65% of patients with RA aged 45–80 years had CAC, compared with 49% in age-matched controls.12 The greatest differences in CAC score between patients with RA and controls are observed in younger individuals aged 45–54 years.13 In a subanalysis of the MESA Study, CAC was increased by a magnitude of 10.5±18 years compared with those without RA, with each year of disease contributing to an extra 0.4 years of CAC above normal ageing.8 Higher CAC scores are seen in patients with longer disease duration and higher clinical disease activity.14
Pattern of CAD in RA
CTCA detects both calcific and non-calcific plaques. In a study of patients with RA who underwent CTCA as part of a routine health evaluation, there was a higher prevalence of multivessel CAD, higher stenosis severity and total plaque burden scores, and a steeper increase in the amount of plaque per age in RA than matched controls.15 Moderate RA disease activity was associated with the presence of both non-calcified and partial calcified plaques in patients ≥55 years, whereas calcified plaque was more common in those with inactive disease. Other studies have confirmed that patients with RA are more likely to have multivessel CAD, with a higher proportion of non-calcified/partially calcified plaques.16 Hence, it is likely that CTCA better captures the spectrum of CAD conferring increased CVD risk in RA compared with CAC scoring alone.
Progression of CAD in RA
Studies evaluating changes in CAC on serial imaging in patients with RA have shown mixed results. Greater progression of CAC than would be predicted based on population-based data from the Heinz Nixdorf Recall Study has been observed in 50% of patients with RA when studied over a 10-year period.17 However, 3-year follow-up data from the MESA Study suggest similar rates of CAC progression for RA to the larger cohort.18 Although CAD progression is non-uniform and influenced by many factors such as age and RA disease duration,19 some longitudinal studies have also found similar rates of progression in those with early versus established RA.18
Effects of RA therapies on CAD
Data evaluating the effects of DMARDs on CAD progression and/or lesion stability using CT in people with RA are limited. In a single-centre study of 150 patients with RA who underwent CTCA and were followed up for a mean 6.9 years, the use of biological DMARDs was associated with lower cardiovascular events in patients with non-calcified or low-attenuation plaque.20 21 Moreover, per-segment plaque analysis in these individuals showed that biological therapy exposure was associated with calcification of previously non-calcified plaques and a lower likelihood of new plaque formation; an effect that was lost on tapering and/or withdrawal of TNFi independent of disease activity. Statin use in this cohort was also found to have a potential beneficial effect on plaque regression and stability by modulating the response to inflammation measured by C reactive protein.20 21
Psoriasis
Plaque psoriasis (also known as psoriasis vulgaris) is a chronic inflammatory skin disease characterised by well-demarcated, salmon-pink/grey plaques covered in silvery scales. Up to 30% of individuals with psoriasis develop psoriatic arthritis, a heterogeneous inflammatory condition that affects joints, entheses, nails and skin. Population-based data have shown that both psoriasis and psoriatic arthritis are associated with increased CVD risk,1 particularly in younger patients.
Risk-prediction scores versus coronary CT in psoriasis
Evidence of the validity of risk prediction tools in psoriasis/psoriatic arthritis is limited. Only one study has directly evaluated clinical CVD risk-prediction scores for detecting subclinical CAD against CAC scanning.22 This cross-sectional study showed a high discordance between the HeartScore 10-year risk algorithm and presence of CAC in 111 patients with chronic severe plaque psoriasis from a Mediterranean population with low-risk CVD, with 25% of patients in non-high-risk categories having CAC. Overall, 17% of patients in low and moderate-risk groups were reclassified as high risk based on CAC score ≥100 as recommended by European Society of Cardiology clinical practice guidelines.
Prevalence of CAC in psoriasis
A meta-analysis of 16 studies involving 3039 patients with psoriasis showed that these patients had a higher prevalence of CAC compared with controls after adjustment for traditional CVD risk factors, with greater risk observed in patients <50 years old.23 An increased prevalence of CAC compared with healthy controls has been shown for both psoriasis and psoriatic arthritis.24
Pattern of CAD in psoriasis
Studies using CTCA have shown that patients with psoriasis have higher total and non-calcified plaque burden, more lesions that are obstructive (stenosis >70%), involve three epicardial vessels or the left main stem, and have high-risk plaque features than individuals without psoriasis.25–28 Psoriasis has also been associated with greater epicardial adipose tissue volume and coronary inflammation measured by perivascular fat attenuation, which are both associated with CVD risk.29
Effects of psoriasis therapies on CAD
Use of biological therapies for psoriasis has been shown to reduce non-calcified plaque burden, lipid-rich necrotic core content and promote plaque stability via increasing calcification.30–32 Improvement in skin psoriasis severity has also been associated with improvement in total and non-calcified plaque burden.27 In another observational study, treatment of psoriasis with biological agents targeting TNFα, IL-12/23 and IL-17 was associated with reduction in coronary inflammation assessed by perivascular fat attenuation index on CTCA after 1 year.33
Systemic lupus erythematosus
SLE is a chronic inflammatory autoimmune disease, which predominantly affects young women aged 15–40 years. Although the clinical manifestations of SLE are varied, accelerated atherosclerosis is frequent and CVD remains the leading cause of death.2 Aside from atherosclerotic CAD, coronary vasculitis is a less common complication of SLE that can result in coronary stenoses, occlusions and aneurysms.
Risk-prediction scores versus coronary CT in SLE
As with other IMIDs, studies have shown clinical risk scores to underestimate the risk of cardiovascular events in SLE, and EULAR does not endorse the use of any particular risk-prediction models because of the paucity of data validating these tools in this target population. When evaluated against CAC, a study showed that the use of both the Framingham Risk Score and the Reynolds Risk Score, which includes C reactive protein, assigned nearly all 21 out of 121 women with SLE and CAC on CT to a low-risk category.34 However, as 20 of these individuals had a CAC score ≥75th percentile for age and sex, adjustment for CAC percentiles lead to 48% reclassification to an intermediate 10-year risk category of >10%. This finding is consistent with another study that demonstrated adjustment of cardiovascular risk status based on CAC score led to 44% reclassification compared to the Framingham Risk Score,35 and other data showing that SCORE classified 36% to a low-risk group who were found to have atherosclerosis of the coronary, carotid or lower extremity arteries.36
Prevalence of CAC in SLE
SLE is an independent risk factor for CAC. There is a higher prevalence of subclinical CAC in patients with SLE compared with controls, with reported odds ratios ranging from 7.7 to 30.37 CAC is seen in 10% of patients with SLE aged 35–39 years, rising to 58% among individuals aged 45–54 years, which is the age group with the most pronounced difference in CAC between SLE and controls.38 Clinical factors in patients with SLE associated with CAC include impaired renal function, duration and disease activity, high anticardiolipin and anti-β2 glycoprotein IgG titres, high C3, positive anti-double-stranded DNA, homocysteine concentration, treatment with corticosteroids and low bone mineral density.3
Pattern of CAD in SLE
In the limited number studies about SLE where CTCA was performed, fibrous plaque has been shown to be the most common plaque type, accounting for up to 85% of total plaque burden.39 There is also a higher prevalence of low-attenuation non-calcified plaque in SLE than controls, and an overall greater non-calcified plaque burden.40 However, most patients with SLE and non-calcified plaque also have some degree of calcified plaque.41 In the latter study of 39 patients, the presence of non-calcified plaque was associated with SLE disease activity and the need for immunosuppression, but not traditional CVD risk factors such as hypertension or cholesterol levels.
Progression of CAD in SLE
Not only does atherosclerotic CAC occur more commonly, and at an earlier age in SLE, but it also progresses faster than in the general population. A longitudinal study of 149 patients with SLE who were followed up for 3.6 years demonstrated progression of CAC scores defined as annualised percentage change relative to baseline in 18.1% vs 1.8% in controls,42 and these findings have been corroborated by other studies. Progression of CAC in SLE has been associated with organ damage assessed by the Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index. Greater atherosclerosis progression has been observed in patients with SLE compared with matched controls in all plaque types except for low-attenuation plaque.39 40
Effects of SLE therapies on CAD
There are limited data on the effects of lipid-lowering and disease-modifying therapies used for SLE on atherosclerotic CAD. The effects of statins on T cell activation and IFN-I responses further complicate the story in SLE. In the randomised placebo-controlled Lupus Atherosclerosis Prevention Study, there was no change in CAC after 2 years following treatment with atorvastatin; however, only 43% of individuals had CAC at baseline.43 A study of 60 individuals with SLE found less progression of CAC in patients randomised to atorvastatin than placebo over 1 year.44 However, while plaque progression is clearly pathological, studies in cohorts of patients without SLE have shown that the plaque-stabilising effects of statins typically reduce non-calcified plaque while increasing the calcific component. Evidence from observational studies and meta-analyses also suggests no effect of mycophenolate mofetil or chloroquine/hydroxychloroquine used for SLE on CAC, respectively.45
Conclusions and future directions
The heightened risk of atherosclerotic CVD in people with IMIDs is well characterised, yet systematically underestimated by conventional primary prevention risk scores. Coronary CT is increasingly being used as an adjunct to clinical risk scores for asymptomatic individuals with borderline indications for statins and other cardiovascular therapies in several clinical settings. However, the role of coronary CT for refining risk prediction in IMIDs is less well studied. Most studies to date have focused on the potential use of CAC score for reclassifying the CVD risk of asymptomatic patients with IMID, with evidence that CAD arises earlier and has faster rate of progression in this target population. The high prevalence of non-calcified plaque in these patients, who are often younger females without traditional CVD risk factors, suggests that CTCA could have an added role for this purpose. CTCA can be used to identify high-risk plaque features and perivascular fat attenuation, which are additional prognostic markers that could be particularly relevant for IMIDs (figure 2). The ongoing SCOT-HEART 2 trial will examine whether using CTCA to screen asymptomatic individuals at risk of CVD in the general population reduces coronary heart disease death or non-fatal MI compared with a clinical risk score approach (NCT03920176). However, only observational data from relatively small cohorts of patients with IMID exist on this topic in the literature, and so there is currently no clear remit for CTCA use in people with IMIDs who do not have chest pain or known coronary disease. Further research is needed to develop robust criteria for identifying people with IMIDs who are most likely to benefit from screening with CAC and/or CTCA, and to determine its overall efficacy in this setting while balancing potential risks including radiation exposure, incidental findings and patient anxiety, as well as the economic implications for healthcare systems.
Ethics statements
Patient consent for publication
Ethics approval
Not applicable.
Acknowledgments
JMT is supported by the Wellcome Trust (211100/Z/18/Z), the British Heart Foundation (BHF) and the BHF Cambridge Centre for Research Excellence (18/1/34212). JHFR is part-supported by the NIHR Cambridge BRC, the BHF, the Engineering and Physical Sciences Research Council, and the Wellcome Trust.
References
Footnotes
Twitter @martapeverelli, @jhfrudd
Contributors All authors contributed to the research and writing of this manuscript.
Funding Wellcome Trust (211100/Z/18/Z); BHF Cambridge Centre for Research Excellence (18/1/34212)
Competing interests None declared.
Provenance and peer review Commissioned; externally peer reviewed.