Association of Metabolic Comorbidities with Pediatric Psoriasis: A Systematic Review and Meta-Analysis.

BACKGROUND: An evident relationship has been shown between psoriasis and metabolic comorbidities. However, the results in pediatric psoriasis vary from study to study, and no meta-analysis exists on the association of metabolic comorbidities with pediatric psoriasis.
OBJECTIVE: To evaluate the association between psoriasis and metabolic comorbidities in pediatric patients.
METHODS: We searched articles published in PubMed, EMBASE, and Cochrane Library databases from inception to April 30, 2019. All observational studies reporting the prevalence of obesity or metabolic comorbidities in pediatric patients with psoriasis were included.
RESULTS: The meta-analysis included 16 unique studies meeting the inclusion criteria. The pooled odds ratios in pediatric patients with psoriasis was 2.40 (95% confidence interval [CI], 1.60~3.59) for obesity (13 studies), 2.73 (95% CI, 1.79~4.17) for hypertension (8 studies), 2.01 (95% CI, 1.09~3.73) for diabetes mellitus (8 studies), 1.67 (95% CI, 1.42~1.97) for dyslipidemia (7 studies), and 7.49 (95% CI, 1.86~30.07) for metabolic syndrome (4 studies).
CONCLUSION: Pediatric patients with psoriasis showed a significantly higher prevalence of obesity, hypertension, diabetes, dyslipidemia, and metabolic syndrome. Adequate monitoring and timely management of metabolic comorbidities should be considered in these patients.

INTRODUCTION

Psoriasis is one of the most common chronic inflammatory skin diseases with a prevalence of approximately up to 2% in adolescents1. Recent studies have suggested that psoriasis is not just an inflammatory skin disease, but it is associated with various systemic diseases2. Increased risk of cardiovascular outcome and metabolic comorbidities, such as diabetes, hypertension, and obesity, has been observed in patients with psoriasis345. Metabolic comorbidity, among several comorbid diseases, is common in developed countries and can be managed through proper lifestyle modification and treatment. Uncontrolled metabolic comorbidity increases mortality by inducing cardiovascular disease6.

The prevalence of psoriasis in children is known to be lower than that in adults17. The prevalence of metabolic comorbidities in children is also low; therefore, accompanying metabolic comorbidities in pediatric psoriasis are easily overlooked. However, the prevalence of psoriasis was reported to increase in adolescence, which is an important period for managing metabolic comorbidities8910.

Several studies observed higher risk of metabolic comorbidities in pediatric groups with psoriasis than in control groups111213 In this regard, active screening for metabolic comorbidities in pediatric patients with psoriasis is recommended, but the level of evidence is low14. Meta-analyses of studies about the association between adult psoriasis and obesity, hypertension, diabetes, dyslipidemia, and metabolic syndrome were performed and confirmed the significant association1516171819, but meta-analysis about the association between pediatric psoriasis and the aforementioned comorbidities has not been performed. In this study, we performed a meta-analysis to evaluate the association of metabolic syndrome including obesity, diabetes, hypertension, and dyslipidemia with psoriasis in pediatric patients.

MATERIALS AND METHODS

This study received an exemption determination from the institutional review board of Seoul National University Hospital (no. 2104-166-1214). Following the Meta-analysis of Observational studies in Epidemiology (MOOSE) guideline, meta-analysis of observational case-control studies, excluding cohort studies, about the prevalence of metabolic comorbidity in pediatric psoriasis was performed20.

Study search and selection

Articles published in PubMed, EMBASE, and Cochrane Library databases from inception to April 30, 2019 were searched. The search was conducted using MeSH terms and free text keywords related to “pediatric,” “psoriasis,” “obesity,” “hypertension,” “diabetes,” “dyslipidemia,” and “metabolic syndrome” (Supplementary Table 1). Reference lists of searched articles were also reviewed.

After excluding duplicate articles, two authors (YEK and SIC) independently screened the title and abstract of articles. Full text was reviewed for articles with title and abstract only which made difficult to evaluate eligibility. Population under 19 years of age (≤18 years of age) was defined as pediatric population. Observational studies, including cross-sectional, case-control, and cohort studies, were included. Studies meeting the following inclusion criteria were finally selected: (1) studies with at least 10 patients with psoriasis vulgaris aged 18 years or below, (2) studies with control group originated from the same database and with at least 10 control subjects, (3) studies measured outcomes for obesity, hypertension, diabetes, dyslipidemia, and/or metabolic syndrome (other outcomes such as overweight, high blood pressure, or insulin resistance were not included), (4) studies with observation period limited to 18 years old or below, and (5) English-written original articles or research letters with adequate information. Studies with more than half of patients have psoriatic arthritis or unable to separate psoriasis vulgaris, only the patient group was excluded. For relevant unpublished studies (studies with abstract only), we attempted to contact with author through e-mail, but studies without sufficient information were excluded.

The Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) diagram of the search and selection of studies is presented in Fig. 1.

Fig. 1

Flowchart of study selection following the PRISMA guideline.

Data extraction and study quality assessment

The following information was extracted from articles by two authors (YEK and SIC).

- General data: First author, year of publication, country, age range, number of cases/controls included, study design

- For comorbidity evaluation: definition of comorbidity, prevalence of obesity/hypertension/diabetes mellitus/hyperlipidemia/metabolic syndrome and information needed to calculate prevalence of obesity/hypertension/diabetes mellitus/hyperlipidemia/metabolic syndrome

The quality of the included studies was assessed by two authors (YEK and SIC) using the Newcastle-Ottawa scale modified by Miller et al.21 (score range 0~11). This scale includes selection (case definition, representativeness of the cases, selection of controls, and definition of controls), comparability, and methods. Publication bias of the included studies was evaluated by visual inspection of funnel plot and Egger's regression test22.

Statistical analysis

Statistical analysis was performed using “meta” package of R version 3.5.0 (R Foundation for Statistical Computing, Vienna, Austria) statistical software. Random effect model was used to calculate pooled odds ratio (OR) with 95% confidence interval (CI). Study heterogeneity was assessed using Cochran's Q test and I2 index. The result with p-value >0.1 from Cochran's Q test or I2 index >75 was considered having substantial heterogeneity and random effect model was applied.

RESULTS

Study selection

A total of 4,532 studies were identified through database search. After excluding 668 duplicate articles and 3,728 inappropriate articles through title and abstract screening, 136 articles were full text reviewed. Articles were excluded for the following reasons: no pediatric data or undistinguishable pediatric data (n=72), no prevalence data (n=7), no detailed data (n=2), no or inappropriate control (n=6), duplicate data source (n=1), studies for psoriatic arthritis (n=2), inappropriate outcome (n=8), inappropriate study design (n=13), and full text unavailable (n=8). A total of 17 studies1112132324252627282930313233343536 met the eligibility criteria; 15 were case-control or cross-sectional studies112345678910111213232425262728293032333435, one was a cohort study36, and one study included both case-control and cohort data31. Meta-analysis for case-control or cross-sectional studies and cohort studies should be performed separately because the outcomes of each are different (prevalence vs. incidence). However, we could not perform meta-analysis for cohort studies because only two met our eligibility criteria. Moreover, some participants in these cohort studies may have become adults during the follow-up period. Finally, 16 studies11121323242526272829303132333435 with 20,676 psoriasis cases and 5,239,197 controls were included for meta-analysis.

Description of the included studies

Characteristics of included studies are summarized in Supplementary Table 2. A total of six studies112629313235 were from the United States, followed by five studies1213253033 from Europe (including France, Germany, Italy, Portugal, and Turkey). All studies were published after 2008. All studies included patients aged 5 to 14 years, and studies including patients aged 0 to 18 years were the most common (n=4)11263031. The design of the 13 studies11122324252627283031323335 was case-control, and that of the other three studies132934 was cross-sectional. For data source, 10 studies12232425262728303335 recruited the participants at medical institutions, and the other six studies111329313234 utilized electronic health record databases. Twelve studies111325262728293031323435 included patients with psoriasis, and the other four studies12232433 included patients with plaque psoriasis. One study12 excluded patients with psoriatic arthritis. Two studies1231 included all five clinical outcomes, and five studies1113252932 included the outcome of obesity, hypertension, diabetes mellitus, and dyslipidemia. One study27 reported the outcome of obesity and metabolic syndrome. Eight studies showed single outcome including obesity (n=5)2324262830, diabetes mellitus, hypertension, and metabolic syndrome (n=1, in each)333435. Half of the included studies1125263031323335 showed high quality with ≥9 points, and the other half of the included studies1213232427282934 showed middle quality with 6 to 8 points according to the modified version of the Newcastle-Ottawa scale (Supplementary Table 3).

Association of pediatric psoriasis and obesity

A total of 13 studies11121323242526272829303132 (20,019 psoriasis cases and 5,219,428 controls) about the prevalence of obesity were identified. Four studies11132932 defined obesity according to the diagnostic code, and nine studies122324252627283031 defined obesity clinically including six studies122426272830 that defined obesity by body mass index ≥95 percentile. Quantitative analysis of 13 articles showed significantly higher odds of obesity in children with psoriasis (pooled OR, 2.40; 95% CI, 1.60~3.59) (Fig. 2A). A random-effects model was applied because a substantial heterogeneity was observed among the 13 articles. Subgroup analysis according to the study design and obesity definition was performed. In 11 studies1112232425262728303132 with case-control study design (17,200 psoriasis cases and 69,886 controls), the odds of obesity were significantly increased in children with psoriasis (pooled OR, 2.14; 95% CI, 1.52~3.01) (Fig. 2B). In nine studies122324252627283031 that clinically defined obesity (2,110 psoriasis cases and 2,122 controls), significantly increased odds of obesity was observed in children with psoriasis (pooled OR, 2.16; 95% CI, 1.48~3.13) (Fig. 2C). In 6 studies122426272830 that defined obesity as body mass index ≥95 percentile (967 psoriasis cases and 615 controls), significantly increased odds of obesity was observed in children with psoriasis (pooled OR, 1.88; 95% CI, 1.06~3.33) (Fig. 2D).

Fig. 2

Association of pediatric psoriasis with obesity. (A) Quantitative analysis of all 13 articles about the association between pediatric psoriasis and obesity. (B) Quantitative analysis of 11 articles with case-control design. (C) Quantitative analysis of nine articles that defined obesity by body mass index. (D) Quantitative analysis of six articles that defined obesity as body mass index ≥95 percentile. OR: odds ratio, CI: confidence interval.

Association of pediatric psoriasis and metabolic comorbidities

A total of eight studies1112132529313233 (18,792 psoriasis cases and 5,218,746 controls) about the prevalence of hypertension were identified. Quantitative analysis of these studies showed that children with psoriasis have significantly increased odds of hypertension (pooled OR, 2.73; 95% CI, 1.79~4.17) (Fig. 3A). Eight studies1112132529313234 (19,325 psoriasis cases and 5,238,391 controls) about the prevalence of diabetes mellitus were identified. Quantitative analysis of these eight studies showed that children with psoriasis have significantly increased odds of diabetes mellitus (pooled OR, 2.01; 95% CI, 1.09~3.73) (Fig. 3B). Seven studies11121325293132 (18,740 psoriasis cases and 5,218,694 controls) about the prevalence of dyslipidemia were identified. Quantitative analysis of these studies showed that children with psoriasis have significantly higher odds of dyslipidemia (pooled OR, 1.67; 95% CI, 1.42~1.97) (Fig. 3C). Four studies12273135 (725 psoriasis cases and 670 controls) about the prevalence of metabolic syndrome were identified. Two studies1235 followed the criteria by de Ferranti et al.37, while one study27 followed the criteria by the International Diabetes Federation38. Quantitative analysis of these studies showed that children with psoriasis have significantly higher odds of metabolic syndrome (pooled OR, 7.49; 95% CI, 1.86~30.07) (Fig. 3D).

Fig. 3

Association of pediatric psoriasis with metabolic comorbidities. (A) Quantitative analysis of eight articles about the association between pediatric psoriasis and hypertension. (B) Quantitative analysis of eight articles about the association between pediatric psoriasis and diabetes mellitus. (C) Quantitative analysis of seven articles about the association between pediatric psoriasis and dyslipidemia. (D) Quantitative analysis of four articles about the association between pediatric psoriasis and metabolic syndrome. CI: confidence interval.

Assessment of publication bias

Funnel plot for studies about hypertension, diabetes, dyslipidemia, metabolic syndrome, and obesity showed no evident asymmetry (Supplementary Fig. 1, 2). Egger's regression test also suggested no significant publication bias (p>0.05).

DISCUSSION

To our knowledge, this is the first meta-analysis on the association between pediatric psoriasis and metabolic comorbidities. This meta-analysis showed that pediatric patients with psoriasis had significantly higher prevalence of obesity, hypertension, diabetes, dyslipidemia, and metabolic syndrome. In addition, significantly increased prevalence of obesity was observed in children with psoriasis regardless of the definition of obesity.

The association of psoriasis with metabolic comorbidities is well reported in adults. Several studies showed strong association of psoriasis with obesity, hypertension, diabetes, dyslipidemia, and metabolic syndrome161718193940. The underlying mechanism of the association between psoriasis and metabolic syndrome is not completely elucidated, but various factors contribute to the association41. Genetic relationship between psoriasis and metabolic comorbidities has also been reported4042. In addition, the concept of “psoriatic march” has been proposed, which means expansion of inflammation from skin to systemic43. The “psoriatic march” hypothesis suggests that locally overproduced pro-inflammatory cytokines in psoriasis lesion are released into the circulatory system and therefore induce systemic resistance, endothelial dysfunction, oxidative stress, and eventually metabolic comorbidities4143. In cases of psoriasis with metabolic comorbidities, proper management and treatment of metabolic comorbidities are crucial to prevent cardiovascular events, the major complication214445. In addition, obesity itself is able to increase the risk and the severity of psoriasis or reduce the psoriasis treatment response, so it should be managed properly464748. Most previous studies, however, are targeted to adults, and studies for children are not sufficient.

Psoriasis in children and adolescents is not uncommon, and a large number of patients with psoriasis who developed psoriasis during childhood experience psoriasis until adulthood1749. As in adults, metabolic comorbidities, such as hypertension, diabetes, dyslipidemia, obesity, and metabolic syndrome, were reported to be increased in pediatric patients with psoriasis75051. The association between pediatric psoriasis and obesity was reported from a previous systematic review. In this study, the association between obesity and psoriasis in adolescents was consistently observed under various definitions of obesity. The associations of pediatric psoriasis with other metabolic comorbidities were not as much clear as that with obesity in all studies5051. Due to the lower prevalence and under-diagnosis of metabolic comorbidities in pediatric population, individual study, particularly about relatively small population, failed to show significant association between metabolic comorbidities and psoriasis505253. However, although not included in the present study, a retrospective cohort study of 29,957 pediatric psoriasis patients reported a significantly higher hazard ratio of hypertension, diabetes, metabolic syndrome, and elevated lipid levels in those patients during the follow-up period36. The risk of metabolic comorbidities was significant independent of obesity36. Taken together with the result of this meta-analysis, the association between metabolic comorbidities and pediatric psoriasis is clear.

Obesity and metabolic comorbidities in childhood have a poor prognosis with multiple complications including cardiovascular disease in adulthood5455565758. In pediatric patients with psoriasis, lipid profile associated with high cardiovascular risks was observed, and increased cardiovascular comorbidities were reported but the frequency is not high2629. Since timely treatment and care of metabolic comorbidities and obesity can reduce the risk of complications, active detection and management of high-risk groups for obesity and metabolic comorbidities among pediatric patients with psoriasis are important14596061. Moreover, treatment with tumor necrosis factor-α inhibitor, interleukin (IL)-12/23 inhibitor, or IL-17 inhibitor was reported to be able to lower the risk of cardiovascular disease in adults; thus, using biologics to pediatric patients with severe psoriasis with metabolic syndrome should be considered62636465. Osier et al.14 proposed an expert consensus recommendation on comorbidity screening guideline in pediatric psoriasis based on studies published up to December 2015 and suggested routine screening for related comorbidities and risk factors identification in pediatric patients with psoriasis. However, evidence level in the study was relative low (Strength of Recommendation Taxonomy level C), and no guideline for management in those patients with metabolic comorbidity was provided. Taken together, further evidence-based guidelines are required to provide adequate metabolic comorbidities screening and management for adolescent patients with psoriasis.

This study has several limitations. First, a total of 16 studies were included in this meta-analysis, but meta-analysis of metabolic comorbidities other than obesity includes less than 10 studies each, which is not adequate. Second, substantial study heterogeneity was observed among the studies. The heterogeneity may result from the difference of each study in subject age, location, severity of psoriasis, and diagnosis criteria of psoriasis, whether diagnosed by dermatologists, medical record, or claims data. For obesity, subgroup analysis was performed according to the study design and diagnosis criteria. In case of other metabolic comorbidities, however, subgroup analysis could not be conducted due to the low number of included studies. Third, there are potential risk of bias such as selection bias and publication bias. Although funnel plot and Egger's regression test suggested no substantial publication bias, potential publication bias cannot be ruled out because of the small number of studies. Finally, this study including only observational studies did not show causality, but only showed the association between psoriasis and metabolic comorbidities. Despite these limitations, this study has significance as the first to present the results of a meta-analysis on the association between pediatric psoriasis and metabolic comorbidities.

This meta-analysis shows that pediatric patients with psoriasis have significantly higher association with obesity, hypertension, diabetes, dyslipidemia, and metabolic syndrome. Dermatologists and pediatricians should be aware of the evident association of metabolic comorbidities with psoriasis in children and should perform early proper investigation and management to prevent disease progress and complications. Further investigations on the casual-relationship and mechanism of the association between pediatric psoriasis and metabolic comorbidities are warranted.