Risk of aortic aneurysm in patients with psoriasis: A systematic review and meta‐analysis of cohort studies

Abstract Background The association between psoriasis and the risk of aortic aneurysm is still unclear. Hypothesis Patients with psoriasis have a higher risk of aortic aneurysm than healthy individuals. Methods PubMed, Embase, and Scopus from inception to 20 July 2019 were searched. We included cohort studies if they reported estimate effects on the risk of aortic aneurysm in patient with psoriasis. We used Newcastle‐Ottawa Scale to evaluate methodology quality of eligible studies. Random‐effect meta‐analyses were used to estimate the overall risk. Subgroup analyses were conducted for analysis of influencing factors. Results After a view of 2207 citations, we included three large cohort studies enrolling 5 706 525 participants in this systematic review. Psoriasis patients have an increased risk of development of aortic aneurysm (hazard ratio [HR]: 1.30, 95%confidence intervals [CI], 1.10‐1.55, I 2 = 53.1%). The risk is not statistically different between patients with severe psoriasis (HR, 1.51, 95%CI, 1.04‐2.19, I 2 = 40.2%) and patients with mild psoriasis (HR, 1.24, 95%CI, 1.08‐1.42, I 2 = 24.1%). The risk was not statistically increased in female patients (HR, 1.55, 95%CI, 0.65‐3.72), patients ≥50 years old (HR, 4.05, 95%CI, 0.69‐23.75, I 2 = 97.3%), and patients with diabetes (HR, 0.97, 95%CI, 0.83‐1.14). Conclusions Current evidence from observational studies suggests that psoriasis increases the risk of aortic aneurysm, and screening of aortic aneurysm might be considered among psoriasis patients.


| INTRODUCTION
Psoriasis is a systemic inflammatory disease that affects an estimated 3% of the US adult population. 1 Patients with psoriasis have an increased prevalence of cardiovascular diseases, including ischemic Xinyu Yu and Xin Feng contributed equally to this study. heart disease, 2 heart failure, 3 peripheral vascular disease, 4 and stroke. 5 The increased risk of cardiovascular events is believed to be associated with the systematic inflammatory pathophysiological mechanisms of psoriasis. 6 Moreover, aortic vascular inflammation detected by 18 Fluorodeoxyglucose (FDG) positron emission tomography/computed tomography (PET/CT) is associated with psoriasis skin disease severity. 1,7 Aortic vascular inflammation plays an essential role in the development and progression of aortic aneurysm (AA). Chronic aortic vascular inflammation is believed to lead to destruction of the aortic media and to vascular smooth muscle cell dysfunction as a result of the release of a range of proteolytic enzymes, such as matrix metalloproteinases and cysteine proteases, oxidation-derived free radicals, cytokines, and related products. 8 Given these foundations, several studies [9][10][11] have explored the relative risk of AA in patients with psoriasis. However, these study findings are conflicting, and whether psoriasis increases the risk of AA is still unclear. Thus, we conducted this systematic review to investigate the risk of AA in psoriasis patients.

| Search strategy
The systematic review followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines. We conducted a systematic search on PubMed, Scopus, and Embase for relevant full-text articles published before July 20, 2019. We did not set any language limitation in the literature search. Search terms (displayed in Electronic supplementary material, ESM) included a combination of keywords relating to psoriasis (eg, "psoriatic arthritis"), aortic vascular inflammation (eg, "vascular inflammation") and aortic aneurysm (eg, "aorta", "aortic aneurysm"). To capture other potentially relevant articles, we manually checked the references of the included literature.

| Inclusion criteria and study selection
To clarify the association between disease severity of psoriasis and the risk of AA, we included observational studies that satisfied the following criteria: studies reported effect estimates on the risk of AA in patients with psoriasis compared to healthy subjects; psoriasis patients undergoing phototherapy, topical therapy, oral-systemic medications or biologic agents for psoriasis treatment were eligible; AA of participants could be thoracic AA or abdominal AA; disease severity of psoriasis in participants could be mild to severe.
Conference abstracts were excluded as their full study reports could not be assessed and their scientific rigor had not been peerreviewed. 12 Case reports and case series were excluded for lack of strict study design. Two investigators (Xinyu Yu and Xin Feng) independently screened identified articles to find eligible studies, and the senior investigator (Shiyi Cao) solved discrepancies in study selection.

| Data extraction
We extracted the following information from eligible studies:

| Quality appraisal
We applied the Newcastle-Ottawa Scale (NOS) to appraise the methodological quality of included studies. The scale is a scoring system covering three perspectives of methodology: selection of study population; comparability; and ascertainment of outcome and exposure. We did not exclude any study on the basis of quality appraisal. Two investigators (Xinyu Yu and Xin Feng) individually assessed the quality of eligible studies, and a senior investigator (Shiyi Cao) solved the discrepancies.

| Statistical analyses
Unadjusted and adjusted hazard ratios (HRs) with corresponding 95% CI were extracted from included studies. We used adjusted HRs to reduce the impact of confounding factors on estimate effects. There was We conducted a meta-regression analysis to test the difference between subgroups. We used Student t test to compare the differences in mean values between two groups, and we conducted Chi-square test to evaluate the significant difference of baseline factors in two groups. All P value were two-tailed and P < .05 was set as the significance level. All analyses were conducted in Stata version 14.0 and forest plots were prepared in R version 3.6.1.

| Study characteristics and quality appraisal
The searching strategy is displayed in Figure 1. In total, 2207 records were identified, and eight articles were assessed for eligibility after screening. Finally, three studies 9-11 with 5 706 525 participants were included in this systematic review. The included studies were scored 7 to 8 points in NOS (Supplementary Table 1). Detailed characteristics of included studies and baseline characteristics of included population were arranged in Table 1 and Supplementary Table 2. Three studies were performed in United States, Denmark, and Taiwan, respectively. Basic features (age, sex, and comorbidity) were significantly different between psoriasis cases and reference population, and we used HRs adjusted by these factors to avoid the interference of them on pooled results.

| Risk of AA in patients with psoriasis
Forest plots of meta-analyses were prepared as shown in Figure 2.
Three studies with 24 864 cases indicated an HR of 1.30 (95%CI,

| Subgroup analysis
According to the subgroup analysis (

| DISCUSSION
This is the first systematic review and meta-analysis to estimate the risk of AA in patients with psoriasis considering the effect of disease severity of psoriasis on the risks. It showed that psoriasis patients are more susceptible to AAs compared to the general population, and the risk does not increase with the severity of psoriasis. Moreover, the association was not significant in female patients, patients ≥50 years old and patients with diabetes.
There was low to moderate statistical heterogeneity within metaanalyses and subgroup analyses. Definitions and inclusion criteria of exposure and outcome were slightly different between included studies (see Supplementary Table 2), and this led to inevitable clinical heterogeneity. Subgroup analyses showed age, gender, and comorbidity with diabetes affect risks of psoriasis. Diabetes is associated with a reduced incidence of AA, 14,15 and it decreased the risk of psoriasis in psoriasis patients in this research.
Psoriasis is an immune-mediated inflammatory disease associated with cardiometabolic comorbidities. 5,16 Studies have confirmed that psoriasis increases the risk of subclinical cardiovascular disease, as evidenced by higher coronary artery calcium 17 and an elevated burden of coronary artery disease. 1,18 Elevated blood inflammatory biomarkers in patients with psoriasis indicate a moderate role of systemic inflammation in pathophysiology of psoriasis. 19 To identify concrete inflammatory lesions, researches detected metabolic activity in vessels and other tissues using 18  and aortic stiffening could also contribute to AA development. 10,24 In patients with psoriasis, increased arterial stiffness is presented 25 and it is associated with systemic inflammation. 26 Aortic stiffening leads to axial stress which then induces and augments processes necessary for AA growth, such as inflammation and aortic wall remodeling. 24 From a preventive point of view, AA screening is strongly suggested recently. 27 In light of present findings, AA screening might When interpreting results of this systemic review, several strengths and limitations should be considered. We conducted subgroup analyses to test the robustness of the results, and we found that age, gender, and comorbidities affect the association. However, we failed to conduct sensitivity analyses and a meta-regression of influencing factors for lack of enough studies. There is moderate heterogeneity in meta-analyses and subgroup analyses, and results of this systematic review should be interpreted carefully.

| CONCLUSION
Our findings indicate a higher incidence of AA in patients with psoriasis compared to the general population, and the risk does not increase with the disease severity of psoriasis. Whether AA screening is effective and feasible in patients with psoriasis still requires further research.

ACKNOWLEDGMENTS
Authors would like to thank all the authors of the original articles. This review was supported by the National Natural Science Foundation of China (NSFC,71603091).