Cardiovascular risk factors and confounders in severe atopic dermatitis: a scoping review

Atopic dermatitis (AD) is a chronic and recurrent inflammatory skin disease that could be associated with cardiovascular diseases (CVDs). Interest in determining whether the chronic inflammation of AD is a risk factor for CVD has been growing in recent years, although there are few studies available on this topic, and the results are quite controversial. The purpose of this scoping review was to understand whether severe atopic dermatitis and its chronic systemic inflammatory process contribute, or not, as a risk factor for the development of CVD. Following the development of a study protocol, a scoping review was carried out based on the structure outlined by Arksey and O'Malley, Levac et al., the Joanna Briggs Institute and the PRISMA-ScR guidelines. PubMed, Cochrane, EMBASE, CINAHL, LILACS, Web of Science, Biblioteca Virtual em Saúde (BVS), Scopus and SciELO were the selected databases. Results: Fifty-two studies were included in this review (18 review studies, 16 cross-sectional studies, 15 cohort studies and three meta-analyses). We found that 36 studies could associate AD with CVD, while 16 studies found no relationship between AD and CVD. Conclusions: It is possible that cardiovascular risk in moderate to severe atopic disease is in confounding factors, and not necessarily because AD and CVDs share the same chronic systemic inflammatory pathway. There is no robust evidence that AD is a risk factor for CVD. To reduce CVD risk in patients with AD, we must pay attention to the risk factors commonly present in these patients and strongly recommend a healthy lifestyle.


Introduction
Atopic dermatitis (AD) is a chronic and recurrent inflammatory disease that affects the skin surface. It has a multifactorial aetiology involving genetic factors and immunological and non-immunological changes. Its prevalence has been increasing in the last 30 years, and approximately 30% of AD patients can be classified as having moderate to severe disease. (Silverwood et al., 2018;Weidinger et al., 2016) AD can have an extremely negative impact on the quality of life of patients and their families, with socioeconomic implications and personal consequences.(Aaron M. Oliveira et al., 2019;Sehra et al., 2008;Silverwood et al., 2018) In recent years, it has been discussed whether AD is an exclusively cutaneous disease or whether it is a systemic inflammatory disease. Some research suggests that AD is predominantly a systemic disease influenced by the T helper (Th) 2 inflammatory axis and its biomarkers and not just a skin-limited disease.
Cardiovascular diseases (CVDs) are the leading cause of death worldwide. Approximately 17 million deaths in 2013 were due to CVD. (Roth et al., 2015) Over the years, many studies have shown an increased risk of CVD in a series of chronic inflammatory diseases, where CVD-related mortality increases by approximately 50%. (Lindhardsen et al., 2016) Endothelial damage, difficulty in vascular remodelling, atherosclerosis and insulin resistance are often related to mild and chronic inflammatory status and ageing. (Ferrucci et al., 2018;Franceschi et al., 2006) Epidemiological studies have confirmed that chronic inflammatory diseases play a fundamental role in endothelial dysfunction and, consequently, in atherosclerosis.
Cardiovascular risk factors (CVRFs), such as metabolic syndrome (MS), obesity, diabetes, chronic renal failure and sleep apnoea syndrome, accelerate arterial ageing and increase the risk of cardiovascular (CV) outcomes. (Castellon et al., 2016;Cuende et al., 2016;Kahlenberg et al., 2013;Manzi et al., 2000) Myocardial infarction (MI) is the most commonly reported CVD in patients with chronic inflammatory disease. The risk of MI among patients with rheumatoid arthritis and severe psoriasis is increased, and an estimated 11,500 CV events per year in the United States are related to these diseases. Armstrong et al., 2013;Lindhardsen et al., 2011) Research on the association of AD with CVD, autoimmune diseases and skin and/or extracutaneous infections has accelerated in recent years, although the correlation is not well established. (Oliveira et al., 2019) Because the correlation between chronic inflammatory disease in severe psoriasis and increased CVD is already well established, currently, increasing attention has been given to severe AD, whose chronic inflammation could be related to an increased risk of CVD and poor outcomes. However, studies have found conflicting results.
To address topics that are controversial, a scoping review can be a good tool, as it is an interactive process. (Arksey & O'Malley, 2005) They are very useful for providing an overview of the available evidence. As there are not very many doubleblind high-quality studies on the association of severe AD with CVD, a meta-analysis cannot be conducted at this time.

Objective
This study aimed to review the scope of publications on the inflammatory mechanisms of severe AD and its relationship with CVD. It is not yet well established whether severe AD and its chronic systemic inflammatory process contribute, or not, as a risk factor for the development of CVD.

Justification
Recently, a possible relationship between CVD and dermatological diseases has received greater attention. An increased prevalence of CV events is observed in patients with severe psoriasis, possibly due to the pro-inflammatory mediators existing in the pathophysiology of skin disease. For this reason, it has been hypothesized that severe AD and its chronic systemic inflammatory state may also be associated with an increased risk of atherosclerosis and CVD, although there are still relatively few studies on this topic.
Recent studies suggest that severe and active AD is associated with a higher risk of CV events and increased mortality. (Hjuler et al., 2015;Silverwood et al., 2018) Patients with AD seem prone to have risk factors for CVD, such as physical inactivity, alcoholism, smoking, obesity, high blood pressure, diabetes and dyslipidaemia. Egeberg et al., 2017;Kantor et al., 2016;Zhang et al., 2015) Several serum biomarkers can be correlated with AD activity and cytokine activation, suggesting a systemic inflammatory disease. In the acute phase of AD, there is a predominance of the T helper cytokine 2 (Th2) axis, with increased activation of cytokines IL4, IL13 and IL31. In the chronic phase, there is polarization to the T helper cytokine 1 (Th1) axis, with increased IL12 and IFN. The severity of AD is significantly correlated with the Th2 axis and IL4 and IL13, in addition to the accumulation of Th22 cells in skin lesions.  Patients with chronic diseases seem to strongly express Th1 and have an increased Th17-mediated inflammatory response. (Ivert et al., 2019) This could explain the association of coronary heart disease and stroke with severe and chronic AD, whereas mild forms of AD are not associated with stroke. (V. Y. F. Su et al., 2014) It is likely that patients with severe and recalcitrant AD may have an increased risk of comorbidities compared to patients with mild AD. (A. M. Drucker & Flohr, 2018) However, there are studies that suggest that AD is not an independent risk factor for CVD. (Andersen et al., 2016;A M Drucker et al., 2016Egeberg et al., 2017;Marshall et al., 2016;Standl et al., 2017;J P Thyssen et al., 2018) Many skin and joint disorders are associated with CV comorbidities because they share aetiological elements. The complexity of the immune system and its role in defending the body requires an arsenal to accurately identify and selectively control inflammatory processes or cells that promote atherosclerosis. It has also been observed that inflammation in skin diseases seems to be an active source of several pro-inflammatory cytokines and chemokines that may predispose patients to CV comorbidities. (Kerkhof & Khamaganova, 2018) Inflammation is known to play a central role in CVD. In the inflammatory cascade, a macromolecular structure called the "inflammasome" stands out. The inflammasome is a signalling platform that detects a variety of triggers of the innate immune system. (Li et al., 2014) For this reason, this complex structure is currently being extensively studied in different epitheliums, including the skin epithelium, and atherogenic inflammasomes are also of great interest. (Palazon-Riquelmeet al., 2018) Secretion of the pro-inflammatory cytokines interleukin-1β and interleukin-18 by inflammasomes triggers an inflammatory cascade that results in a strong immune response and can culminate in the progression of diseases such as atherosclerosis, ischaemic injury, cardiomyopathy and skin diseases such as psoriasis and AD. (Li et al., 2014)

Methodology
In this study, a scoping review was carried out. The methodology was based on the structure outlined by Arksey and O'Malley and the methodological improvement made by Levac et al. and the Joanna Briggs Institute. (Arksey et al., 2005;Levac et al., 2010;Peters et al., 2020) STEP 1: Formulation of the research question.
This scoping review was guided by the question: 'Does chronic inflammation in severe atopic dermatitis make it a risk factor for cardiovascular disease?'

STEP 2: Identification of the relevant studies
The descriptors used in this research to identify the eligible studies are as follows: • ("Dermatitis, Atopic" OR "Atopic Dermatitis" OR "Atopic Eczema") AND ("cardiovascular diseases" OR stroke OR "metabolic syndrome" OR "coronary disease" OR "myocardial infarction" OR atherosclerosis).

STEP 3: Study Selection
The review process consists of 3 levels of screening: (1) title review, (2) summary review and (3) full text review. For the first level of screening, two researchers independently screened the title of the 849 eligible articles found with the search strategy. Subsequently, the abstracts of all 712 citations retrieved for inclusion were screened against a set of minimum inclusion criteria. In the third stage, the two researchers independently evaluated the 52 articles selected from the analysis of the abstracts in full text to determine whether they met the inclusion/exclusion criteria.

STEP 4: Data mapping
The results obtained after analysing the full-text articles and the excluded data were reviewed a second time, and any disagreements about the study's eligibility in the text review phase were resolved through discussion among the researchers. The search results were imported into the Mendeley® quote manager and grouped into a single library.

STEP 5: Collection, summarization and reporting of the results
Critical assessment of the risk of bias was carried out by researchers FWR and ELJ. The PRISMA-ScR (Preferred Reporting Items for Systematic reviews and Meta-Analyses) guidelines were followed to report the findings of this scoping review ( Figure 1). (Tricco et al., 2018).

Results
In this review, 52 studies were included (18 review studies, 16 cross-sectional studies, 15 cohort studies and three meta-analyses).
According to the analysis of the results, 36 studies could associate AD with CVD (Table 1, Table 2, Table 3 and Table 4).  Atherosclerotic plaques in patients with AD were significantly higher. This suggests that AD is associated with an increased prevalence of disease coronary artery.

Discussion
The inflammatory nature of AD could be responsible for an association with CVD, as in other chronic inflammatory diseases such as psoriasis and rheumatoid arthritis, whose evidence is well established. However, alcoholism, smoking, dyslipidaemia, sleep disorders, metabolic syndrome, mood disorders, obesity, overweight, physical inactivity, and diabetes, which are known to be factors predisposing to CVD, are present in most AD patients and are considered confounding factors. (Ali et al., 2018) Most comorbidities associated with AD are apparently caused by the interaction between genetic predisposition, environmental exposure, chronic inflammation due to AD, medication use and lifestyle.  In fact, the heterogeneity found in the studies can be attributed to variations in the studied populations, the study design, the follow-up time and adjustment for confounding factors. These factors seem to be of great relevance for the interpretation of the results of the studies and the scientific data analysed.
Under a critical epidemiological view, it is possible to explain, in large part, the possible associations between AD and CVD with respect to these confounding factors. The literature points out that AD is associated with sleep disorders and obesity, and both are strongly related to CV events.(Aaron M. Drucker & Harvey, 2019) Some studies concluded that being male is associated with AD and an increased risk of stroke and myocardial infarction (MI), but it should be considered that in men, the occurrence of factors such as smoking, alcoholism, and sleep disorders is more prevalent. (Yuan et al., 2018) In addition, some studies have identified AD through questionnaires and self-reports. This fact could lead to a risk of bias, which could certainly lead to an incorrect classification of AD, psoriasis and other pruritic diseases of the skin.
Misclassification bias may be responsible for an increased risk of CV outcomes in atopic patients, as those with severe disease are often defined by receiving systemic therapies, and some of these therapies increase the CV risk, such as systemic corticosteroids. (Wei et al., 2004) The time of onset of AD, whether in childhood or adulthood, has been little reported in the literature, but it can be a significant problem due to the timing of progression of the disease and the timing of the exposure to CV risk factors or medications. (Ascott et al., 2019) In an arbitrary way, in many studies analysed in this review, they considered that an HR (hazard ratio) or (odds ratio) less than two (2), which is a "small" variation, can be explained only by confounding factors or biases in the different types of study designs and their heterogeneity. (Cribier, 2019) This increases the risk of a false positive. Studies that intend to assess the risk of CV events and that have fewer than two years of follow-up of patients are not considered adequate to assess the risk of more serious outcomes, such as MI, stroke and mortality. This is due to the time required for the progression of this outcome in the studied groups, which evolves insidiously and probably requires more than 18 months to 24 months. Therefore, studies with a longer follow-up time can present more robust data to be analysed.
There is a possibility that studies that found a positive association between AD and CVD are making a type I error or an alpha (α) error because they do not take into account confounding factors. In statistics, an alpha error or type I error occurs when a hypothesis is rejected when it is true. This can occur in research when a different result is found randomly or due to the confounding factors existing in the studied population. Thus, what could be causing CVD would be CV risk factors, which are often present in patients with moderate to severe AD. This does not make it an isolated risk factor for CVD but a possible residual risk factor or risk marker.
The study design that could best answer the question "Is severe atopic dermatitis a risk factor for cardiovascular disease?" would be two study groups, the first group of which would be patients with MS and severe AD and the second group would be with MS without severe AD. Both groups were treated for MS and followed for three to five years. If the group of patients with MS and severe AD had more fatal CV outcomes, it could be concluded that severe AD is an independent risk factor for CVD. Thus, it would define the burden of AD in CVD, which could be considered a neglected residual risk factor.
The CV mortality rates, despite having suffered a significant reduction in recent years, are still considered high, possibly due to the residual risk of CV events. (Hermans & Fruchart, 2010) Residual risk is comprised of modifiable risk factors, such as lipid changes, diabetes, hypertension, physical inactivity and smoking, and nonmodifiable risk factors, such as age, sex, previous CVD and genetic factors. (Thomas et al., 2018) Since the major components of residual risk are related to dyslipidaemia and hypertension, changes in lifestyle associated with usual drug therapy can strongly contribute to the reduction of CVD and fatal outcome events. (Hermans & Fruchart, 2010) Recent studies suggest, although still with conflicting results, that AD-related comorbidities increase morbidity and mortality in paediatric and adult patients with both psoriasis and AD. Many of the pathophysiological mechanisms of these comorbidities associated with AD are still not fully understood. However, AD itself and the associated lifestyle contribute to mechanisms to trigger such comorbidities. It is possible that adequate and early treatment of AD, in addition to reducing risk factors, can contribute to the prevention of these comorbidities.  What is the importance of these studies for the patient and the doctor? For patients with moderate to severe AD, the evidence points to the need to map risk factors for CVD. Age-appropriate cardiometabolic screening should be performed. There is insufficient evidence to state that the treatment of systemic inflammation in AD would have an effect on the patient's cardiometabolic profile. It is likely that the appropriate treatment of AD, with an improvement of symptoms, may, in turn, improve the quality of sleep, mood and the ability to tolerate physical exercise, thus being able to contribute to a reduction in the general CV risk.(Aaron M. Drucker & Harvey, 2019) The severity of atopic disease and the patient's lifestyle are closely associated, as well as comorbidities, and perhaps more attention should be given to those patients with moderate to severe disease. (Vestergaard, 2019) The origin of these comorbidities is probably multifactorial, including rupture of the skin barrier, immunological dysregulation, extensive symptoms and iatrogenic complications. Many of these comorbidities are directly related to the underlying severity of AD and to inadequate disease control, and they could be avoided or mitigated by optimizing AD management.(Jonathan I Silverberg, 2018) Due to the prevalence of adults with eczema or AD, any reduction in the risk of CVD is important from the point of view of public health. (Fujiyoshi, 2019) Therefore, physicians must rethink whether interventions in this group of patients could prevent fatal outcomes.
Evidence from new treatment perspectives that could contribute to reducing CV events would be an important advance to guide the clinical conduct of health professionals. (Ingram, 2018) Epidemiological studies aim to reduce bias and confounding effects, but the statistical associations have not yet been sufficient to prove the causality of the relationship between AD and CVD. It is possible that monitoring patients longitudinally to determine the impact of new therapeutic approaches on comorbidities associated with AD, in addition to its cutaneous manifestations, might raise new hypotheses about the mechanisms of the associations between these diseases. (Brunner, Silverberg, et al., 2017) Although there are contradictory results, together, these findings suggest that an increase in risk factors for CVD and poor health behaviour, lifestyle factors such as smoking, drinking and physical inactivity, may be important factors for cardiovascular events instead of considering only the systemic inflammation associated with AD.
Patients should be encouraged to modify these habits and acquire a healthier lifestyle. The growing body of evidence for potential comorbidities associated with AD and its related systemic immune activation reflects the disease burden and supports the concept of AD as a systemic disease, reinforcing the importance of considering the general health of patients, not just their skin disease. (Oliveira et al., 2019)

Conclusion
It is possible that cardiovascular risk in moderate to severe atopic disease is in confounding factors, not necessarily because AD and CVD share the same chronic systemic inflammatory pathway. There is no robust evidence that AD alone is a risk factor for CVD.
For future studies, it is necessary to identify and adjust the confounding factors and mediators to address this unanswered question: whether the inflammatory burden of AD contributes to CVD independently. This would help to outline the contribution of the increased burden of CV risk factors and the increased risk of CVD conferred by AD. (Ascott et al., 2019) For now, perhaps the main objective is to improve the general health status of patients with AD.

Strengths and limitations of our study:
►To address topics that are still controversial, such as the inflammatory mechanisms involved in severe atopic dermatitis and cardiovascular diseases, a scoping review is a good first approach.
►This scoping review included a large volume of literature on a broad topic, and the results highlight the possibility of cardiovascular risk in moderate to severe atopic disease is in confounding factors, and not necessarily because AD shares the same chronic systemic inflammatory pathway with CVD.
►The scoping review approach has some limitations, as it does not formally assess the quality of the evidence and generally collects information from a wide variety of study models and methods.
►Scoping reviews do not provide a synthesized result or answer to a specific question, but they do provide a more general view of the available literature.

Patient and Public Involvement
No patients were directly involved.

Ethics and Dissemination
This study does not require ethical approval.
Author's Contribution FWR conducted the literature review, data collection, data analysis and writing of the manuscript. ELJ was responsible for the idea and design of the study and conceived the research question. He was involved in data verification, student supervision, and editing of the manuscript. Both authors had full access to all of the data in this study and can take responsibility for the integrity of the data and the accuracy of the data analysis.
Funding This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.