Author + information
- Received June 9, 2017
- Revision received August 15, 2017
- Accepted August 16, 2017
- Published online February 5, 2018.
- Joshua P. Rivers, MSa,
- Tiffany M. Powell-Wiley, MD, MPHa,
- Amit K. Dey, MDa,
- Justin A. Rodante, PA-Ca,
- Jonathan H. Chung, BAa,
- Aditya A. Joshi, MDa,
- Balaji Natarajan, MDa,
- Aparna P. Sajja, BA, BSa,
- Abhishek Chaturvedi, MDa,
- Anshuma Rana, MDa,
- Charlotte L. Harrington, BAa,
- Heather L. Teague, PhDa,
- Benjamin N. Lockshin, MDb,
- Mark A. Ahlman, MDc,
- Jianhua Yao, PhDc,
- Martin P. Playford, PhDa,
- Joel M. Gelfand, MD, MSCEd and
- Nehal N. Mehta, MD, MSCEa,∗ ()
- aNational Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland
- bDermAssociates, Silver Spring, Maryland
- cDepartment of Radiology and Imaging Sciences, National Institutes of Health Clinical Research Center, Bethesda, Maryland
- dDepartment of Dermatology, Hospital of the University of Pennsylvania, Philadelphia, Pennsylvania
- ↵∗Address for correspondence:
Dr. Nehal N. Mehta, Section of Inflammation and Cardiometabolic Diseases, National Heart, Lung, and Blood Institute, 10 Center Drive, Clinical Research Center, Room 5-5140, Bethesda, Maryland 20892.
Objectives The authors sought to examine the relationship between visceral adipose tissue (VAT) and vascular inflammation (VI) by 18F-Fluorodeoxyglucose (18F-FDG) positron-emission tomography (PET)/computed tomography (CT) in psoriasis (PSO). Furthermore, we evaluated whether treatment of PSO modulated VAT and VI.
Background PSO, a chronic inflammatory skin disease, is associated with VI by 18F-FDG PET/CT and increased cardiometabolic risk including adipose tissue dysregulation. Recently, VI was associated with future cardiovascular events; however, the relationship of visceral and subcutaneous adiposity with VI in PSO has yet to be evaluated.
Methods Consecutive PSO patients (N = 77) underwent 18F-FDG PET/CT scans to measure VI and abdominal adiposity. A subset of PSO patients with severe skin disease was scanned at 1 year following PSO treatment (N = 13).
Results The cohort was middle aged (51.8 ± 12.6 years), predominantly male (n = 44, 57%), had low cardiovascular risk by Framingham 10-year risk (median 4 years [interquartile range (IQR): 2 to 7 years]), and mild-to-moderate skin disease (5.2 [IQR: 3.0 to 8.5]). PSO disease severity associated with VAT (β = 0.33; p = 0.004) beyond SAT (β = 0.30; p = 0.005). VAT (β = 0.55; p < 0.001), but not SAT (β = 0.15; p = 0.11), associated with VI beyond cardiovascular risk factors. We followed a subset of severe PSO patients treated aggressively for PSO and observed improvement in PSO severity and VAT, which was associated with an improvement in VI at 1 year beyond cardiovascular risk factors (β = 0.53; p = 0.049).
Conclusions Volume-based CT measurement of VAT may capture metabolic risk associated with VI compared to subcutaneous adipose tissue in PSO. PSO treatment associated with a decrease in VAT as well as decrease in VI suggesting VAT as a relevant biomarker related to VI in PSO.
- cardiometabolic disease
- cardiovascular disease
- 18F-FDG PET/CT
- vascular inflammation
- visceral adiposity
Psoriasis, an immune-mediated, chronic inflammatory skin disease, affects up to 3% of the adult U.S. population. According to the National Psoriasis Foundation, approximately 1 million Americans suffer from moderate to severe psoriasis, a third of which are treatment naïve (1). Although the psoriatic plaque may appear to be limited to the skin, its effects may be far-reaching and systemic leading to increased obesity and cardiometabolic disease risk (2). Patients with psoriasis also have an increased risk of developing inflammatory atherogenesis with a subsequent increase in the incidence of early cardiovascular events including myocardial infarction (3), stroke (4), and premature cardiovascular death (5). Therefore, psoriasis serves as a pragmatic model to understand obesity and the development of subclinical inflammatory atherogenesis in humans.
As a cardiovascular disease risk factor, obesity has been targeted as a major modifiable cardiometabolic parameter (6). Thus, there are important implications for dissecting potential mechanisms underlying the contribution of obesity to cardiovascular morbidity and mortality especially in inflammatory disease states such as psoriasis. Obesity is traditionally assessed by anthropometric measures such as body mass index (BMI) and waist-to-hip ratio, but these metrics do not effectively capture cardiometabolic disease risk, particularly in populations of varying ethnic backgrounds (7). Moreover, direct assessment of abdominal adiposity through computed tomography may provide a more accurate evaluation of adipose dysfunction and complement information obtained through BMI and/or waist-to-hip ratio. Furthermore, the distinction of metabolic activity between visceral and subcutaneous adipose compartments, suggests a differential interplay with cardiometabolic risk and further implicates the utility of direct adipose quantification (8).
Vascular inflammation by 18F-Fluorodeoxyglucose (18F-FDG) positron-emission tomography (PET)/computed tomography (CT) provides a reliable, prognostic marker of subclinical vascular disease. Previous studies examining vascular inflammation by 18F-FDG PET/CT in psoriasis have provided important proof-of-concept for use of FDG in detecting vascular inflammation thought to be influenced by remote skin disease (9). Furthermore, we recently showed a dose-dependent relationship between improvement in psoriasis skin disease severity and improvement in vascular inflammation at 1 year demonstrating a potentially shared mechanism between remote skin inflammation and the development of vascular disease (9,10). Moreover, vascular inflammation by 18F-FDG PET/CT has been shown to precede subclinical atherosclerosis and predict the risk for development of future cardiovascular events (11), thereby providing an opportunity to better understand the evolution of vascular disease from its early stages.
Given the apparent interplay of psoriasis, obesity, and cardiovascular disease, we hypothesized the following: 1) visceral adiposity would associate with vascular inflammation independent of cardiovascular risk factors whereas subcutaneous adiposity would not; and 2) improvement in visceral adiposity would associate with improvement in vascular inflammation at 1 year of psoriasis intervention in a subset of psoriasis patients with severe skin disease.
A total of 77 psoriasis patients were consecutively recruited between January 23, 2013, and September 18, 2014, in an observational longitudinal cohort study design (Online Figure 1). These patients were recruited as a part of an ongoing cohort study to understand the relationship between psoriasis and cardiometabolic diseases. The primary outcome for our study was aortic vascular inflammation by 18F-FDG PET/CT and the exposure of the study was adipose depots measured via CT (Figure 1). The details of study participants, inclusion/exclusion criteria, primary outcomes, specifics of imaging procedures, and statistical analyses are presented in the Online Appendix.
Entire sample and sample stratified by psoriasis skin disease severity
The cohort was comprised of 77 consecutively recruited patients who were predominantly Caucasian (n = 68, 88%), middle-aged (51.8 ± 12.6 years), and predominantly male (n = 44, 57%) (Table 1). Patients had a low cardiovascular risk by Framingham 10-year risk (median 4.0 [interquartile range (IQR): 2.0 to 7.0]), despite being obese (30.1 ± 5.3 kg/m2), having mild insulin resistance (3.2 [IQR: 2.0 to 5.0]), and more than one-half having a history of dyslipidemia (n = 50, 65%). The prevalence of diabetes mellitus was 13% with mostly normal fasting glucose levels (102.5 ± 21.5 mg/dl). The cohort had a mild/moderate psoriasis severity as measured by psoriasis area and severity index score (5.2 [IQR: 3.0 to 8.5]), a mean psoriasis duration of 21.3 ± 15.5 years, and high-sensitivity C-reactive protein (1.8 mg/l [IQR: 0.8 to 4.2 mg/l]) was also elevated. Approximately one-half of the patients were on systemic or biologic therapy for psoriasis (n = 31, 40%) and approximately one-half were on lipid therapy (mostly statins) (n = 31, 40%). Furthermore, when stratified by psoriasis skin disease severity, patients with severe psoriasis had significantly higher BMI, insulin resistance, visceral adiposity volume, and vascular inflammation by 18F-FDG PET/CT.
Adiposity, vascular inflammation, and psoriasis severity
To understand dose-response effects of visceral adiposity on cardiometabolic risk factors, we stratified the sample into tertiles by visceral adiposity and found a consistent trend across all 3 tertiles (Online Table 1). Visceral adiposity was observed to be higher in patients who were older, male, and obese. Patients with higher visceral adiposity also had increased prevalence of hypertension, hyperlipidemia, higher cardiovascular risk by Framingham 10-year risk, more insulin resistance by homeostasis model assessment of insulin resistance (HOMA-IR), and more severe psoriasis. In univariate analyses, visceral adiposity directly associated with cardiovascular risk by Framingham 10-year risk, insulin resistance, anthropometric measures such as BMI and waist-to-hip ratio, psoriasis area and severity index score, and vascular inflammation by 18F-FDG PET/CT (Online Table 2). Psoriasis disease severity associated with visceral adiposity volume (β = 0.33; p = 0.004) beyond subcutaneous adiposity (β = 0.30; p = 0.005) whereas subcutaneous adiposity did not (β = 0.09; p = 0.43) (Figures 2A and 2B).
Subcutaneous adiposity (β = 0.37; p = 0.001) as well as visceral adiposity volume (β = 0.71; p < 0.001) associated with vascular inflammation in univariate analyses (Table 2, Figures 2C and 2D). Visceral adiposity remained significantly associated with vascular inflammation after adjustment for cardiovascular risk by Framingham 10-year risk, subcutaneous adiposity, type 2 diabetes mellitus, and lipid therapy (β = 0.55; p < 0.001), whereas the relationship with subcutaneous adiposity was no longer significant following adjustment for the same covariates (β = 0.15; p = 0.11). When comparing between patient groups with varying degrees of skin disease severity, subcutaneous (β = 0.39; p = 0.002) and visceral adiposity (β = 0.70; p < 0.001) were associated with vascular inflammation in univariate analysis for the mild-to-moderate disease group; however, in the severe disease group, only visceral adiposity (β = 0.69; p = 0.003) was associated with vascular inflammation (Table 2). After adjusting for Framingham 10-year risk, subcutaneous adiposity, type 2 diabetes mellitus, and lipid therapy, visceral adiposity remained associated with vascular inflammation for both mild-to-moderate (β = 0.61; p < 0.001) and severe (β = 0.55; p = 0.03) disease groups. However, the relationship between subcutaneous adiposity and vascular inflammation in the mild-to-moderate disease group was attenuated after adjustment for Framingham 10-year risk, visceral adiposity, type 2 diabetes mellitus, and lipid therapy (β = 0.14; p = 0.16). In the overall cohort, sensitivity analysis after removing those on systemic or biologic therapy yielded consistent results for the relationship of visceral adiposity with vascular inflammation (β = 0.72; p < 0.001). We also tested for sex-visceral adiposity interaction and found no significant interaction (p > 0.10). Finally, there were no differences in the relationship between subcutaneous and visceral adiposity on vascular inflammation by sex (p > 0.10).
To contextualize whether visceral adiposity provided incremental value beyond traditional anthropometrics in association with vascular inflammation, we performed nested modeling using visceral and subcutaneous adiposity as well as traditional anthropometric measures such as BMI and waist-to- hip ratio. Visceral adiposity provided incremental data beyond traditional risk factors, subcutaneous adiposity, BMI, and waist-to-hip ratio in nested models (Online Table 3).
Effect of treatment of psoriasis on the relationship between visceral adiposity and vascular inflammation
At 1-year follow-up examination, the cohort had a slight decrease in Framingham 10-year risk (baseline [6 (IQR: 4 to 10)] vs. 1-year [4 (IQR: 2 to 10)]; p = 0.16) although not statistically significant (Online Table 4). Moreover, there was a small reduction in high-sensitivity C-reactive protein (p = 0.01) which persisted even after removal of the most severe responder (p = 0.02). Psoriasis skin disease severity (70%; p = 0.004) decreased significantly at 1 year accompanied by an 8.7% decrease in subcutaneous adiposity (baseline 27,186.9 ± 14,524.7 cm3 vs. 1-year 24,831.3 ± 12,602.5 cm3; p = 0.02), a 10.3% decrease in visceral adiposity (baseline 26,165.0 ± 8,405.5 cm3 vs. 1-year 23,470.7 ± 8,199.6 cm3; p = 0.003), and a 9.6% reduction in vascular inflammation (baseline 1.98 ± 0.38 vs. 1-year 1.79 ± 0.31; p = 0.02) over 1 year (Online Figure 2). Furthermore, improvement in visceral adiposity was associated with an improvement in vascular inflammation after adjustment for traditional cardiovascular risk factors, subcutaneous adiposity, type 2 diabetes, lipid therapy, systemic, and/or biologic psoriasis treatment (β = 0.53; p = 0.049). However, this was not true for the improvement in subcutaneous adiposity (β = 0.39; p = 0.27) (Table 3).
Using a longitudinal cohort study design in a well-phenotyped sample of psoriasis patients with 1-year follow-up, we demonstrate these findings: 1) psoriasis severity appears to have a dose-response relationship with increasing cardiometabolic disease risk factors, visceral adiposity, and vascular inflammation; 2) visceral adiposity is associated with vascular inflammation beyond cardiovascular risk factors compared to subcutaneous adiposity, BMI, and waist-to-hip ratio; and 3) a decrease in visceral adiposity associates with an improvement of vascular inflammation following 1 year of psoriasis therapy. Collectively, these findings suggest that visceral adiposity plays a role in vascular inflammation in psoriasis beyond BMI.
Obesity has been shown to be highly prevalent in chronic inflammatory disease states such as psoriasis. In fact, multiple epidemiologic studies have shown a dose-dependent relationship between increasing obesity and psoriasis skin disease severity (12). This relationship may be driven in part by psoriatic plaques directly by the release of inflammatory cytokines observed in psoriasis (13), and this relationship may be bidirectional whereby visceral adiposity may produce inflammatory mediators which worsen psoriasis. In addition to this, numerous studies in the past have also shown an increased risk of atherosclerotic disease in people with psoriasis (14). Recent findings have indicated psoriasis to be associated with cardiometabolic dysfunction including impaired high-density lipoprotein function, insulin resistance, and obesity (12,15). Indeed, psoriasis has also been shown to have more subclinical atherosclerosis as compared to controls which is evidenced by greater vascular inflammation by 18F-FDG PET/CT (9), higher coronary artery calcium score (16), and increased coronary artery plaque burdens by coronary computed tomographic angiography (17). In addition to the chronic systemic inflammation observed in psoriasis, increased rates of obesity also contribute to some of the increased cardiovascular risk seen in patients with psoriasis (18).
Obesity has been a well-known characterized risk factor for cardiovascular disease (7). Moreover, the physiological distinction between visceral and subcutaneous adiposity has emerged as an important determinant in assessing cardiovascular disease risk. Visceral adiposity has been shown in the past to be highly metabolically active through production of peptides and nonpeptides which may play a role in regulation of cardiovascular homeostasis (19). Inflammatory visceral adipose tissue dysregulation can alter the immune cell and adipokine profile exacerbating endothelial dysfunction (20). Leptin, for example, is a proinflammatory adipokine which promotes expression of other adipocytokines such as tumor necrosis factor (TNF)-α, interleukin (IL)-2, IL-6, and monocyte chemoattractant protein-1, all of which are prothrombotic and promote vascular inflammation (21). Ghrelin, a peptide hormone, has been shown to be protective against endothelial dysfunction and subsequent vascular inflammation via increasing nitric oxide synthesis and reducing vascular cell adhesion. Although not directly related to adipocyte function, ghrelin’s release is inhibited, in part, by leptin (22). In addition to ghrelin, adiponectin, an anti-inflammatory adipokine, has been shown to have protective effects on the endothelium via inhibition of macrophage activity, reduction in endothelial nuclear factor kappa B, and reduction in TNF-α and interferon-γ (23).
The distinctive metabolic profile of visceral adiposity indeed contributes toward increased cardiovascular disease risk, specifically in inflammatory disease states; and thus, it should be evaluated independently (24). CT is one of the most accurate ways in which adiposity can be evaluated (25). In the past, visceral adiposity by CT has been shown to be associated with subclinical cardiovascular disease as evidenced by vascular inflammation measured by 18F-FDG PET/CT independent of cardiovascular risk factors in a cohort of patients with history of cancer, but lacking active disease (11). Furthermore, visceral adiposity has been shown to be an independent predictor of cardiovascular events such as stroke, unstable angina, myocardial infarction, heart failure, and cardiovascular death (11). In our study, visceral adiposity was associated with cardiovascular risk factors, and provided the largest incremental value beyond cardiovascular risk factors for vascular inflammation (beyond subcutaneous adiposity, BMI, and waist-to-hip ratio) suggesting that visceral adipose tissue confers metabolic risk beyond conventionally defined obesity.
Previous studies have highlighted the impact of biologic treatment of psoriasis on decreasing the incidence of myocardial infarction (26). In addition to this, treatments such as anti-TNF and anti-IL 12/23 therapies reduced vascular inflammation by 18F-FDG PET/CT (10) and reduced progression of coronary artery disease as assessed in observational studies by coronary artery calcium score in psoriasis (27). In this study, we showed that a 10.3% decrease in visceral adiposity was associated with a 9.6% improvement in vascular inflammation after 1 year of psoriasis treatment. This relationship remained robust even after adjustment for traditional cardiovascular risk factors and provide compelling preliminary evidence that pathways associated with visceral adiposity may play a role in vascular inflammation; however, larger studies are needed to confirm these findings.
The observational nature of our study renders it prone to error by confounding covariates; a common feature of all observational studies. Additionally, our small sample size is also a limitation; however, this is the largest effort to date evaluating adipose depots as they relate to vascular inflammation. Furthermore, the significance and strength of the noted relationships are robust and persisted beyond adjustments for cardiovascular risk. Admittedly, we do recognize that the relationship of subcutaneous adiposity with vascular inflammation may have lacked significance due to reduced power. Studies of this nature should be performed in larger more generalizable cohorts to confirm the results presented in this study. We also could not evaluate cardiovascular events but instead used vascular inflammation by 18F-FDG PET/CT to understand modulation of cardiovascular disease risk. 18F-FDG PET/CT, however, is a reliable and robust surrogate of future cardiovascular events (11). Finally, we did not do any direct adipose tissue assessment in this study, as we have done in the past (20). Future studies should evaluate the relationship of visceral adiposity with vascular inflammation in a larger, generalizable cohort further evaluating mechanistic links between adipose tissue depots and subclinical atherosclerosis. The findings from our observational study should subsequently be confirmed by performing randomized clinical trials to assess the impact of novel psoriasis treatment modalities on modulation of adiposity as well as cardiometabolic disease. In addition to visceral adiposity volume, localized adipose inflammation (28) and a more specific evaluation of adipokine profile (29) have been of interest in prior studies. The quantification of adipose inflammation may further reveal mechanistic relationships with visceral adiposity and vascular inflammation.
Visceral adiposity was associated with cardiovascular disease risk factors including vascular inflammation in psoriasis. Moreover, this relationship with vascular inflammation remained robust beyond cardiovascular risk factors for visceral compared to subcutaneous adiposity. Furthermore, visceral adiposity provided additional value in association with vascular inflammation over subcutaneous adiposity as well as current measures of obesity such as BMI and waist-to-hip ratio beyond cardiovascular risk factors. Following treatment of psoriasis, improvement in vascular inflammation and visceral adiposity suggested that modulation of visceral adipose volume may play a role in reducing vascular inflammation and subsequent cardiovascular disease risk in a chronic inflammatory state.
COMPETENCY IN MEDICAL KNOWLEDGE: Psoriasis is associated with increased risk of obesity. As such, measures of adiposity such as BMI and waist-to-hip ratio are important to assess when caring for patients with psoriasis. This paper underscores the importance of visceral adiposity as a risk factor for vascular diseases. In addition, the reduction of visceral adiposity after psoriasis treatment may in part play a role in amelioration of vascular inflammation at 1 year. Moreover, the association of psoriasis severity with visceral but not subcutaneous adiposity suggests that systemic inflammation in psoriasis differentially impacts adipose depots. Furthermore, these data support assessing general measures of anthropometrics when providing clinical care for patients at high risk for vascular diseases such as those with psoriasis.
TRANSLATIONAL OUTLOOK: Both visceral and subcutaneous adiposity were associated with vascular inflammation in our study. However, when adjusted for cardiovascular risk factors, the relationship was attenuated for subcutaneous adiposity. These findings suggest that studies should focus on developing mechanistic understanding of the visceral adipose depot in psoriasis given the metabolic activity and contribution to atherogenesis from this depot remains poorly understood.
Dr. Lockshin has received fees from Lilly, Novartis, Janssen, and Abbott. Dr. Gelfand was supported by an NIAMS grant (K24-AR-064310); has received consultant fees from Coherus (DSMB), Dermira, Janssen Biologics, Merck (DSMB), Novartis Corp., Regeneron, Sanofi, and Pfizer Inc.; has received research grants (to the Trustees of the University of Pennsylvania) from Abbvie, Janssen, Novartis Corp., Regeneron, Sanofi, Celgene, and Pfizer Inc.; has received payment for continuing medical education work related to psoriasis that was supported indirectly by Lilly and Abbvie; and is a co-patent holder of Resiquimod for treatment of cutaneous T cell lymphoma. Dr. Mehta has received funding from the National Institutes of Health Intramural Research Program (Z01 HL-06193); is a full-time U.S. Government employee; and has received research grants from Abbvie, Janssen, Novartis Corp, and Celgene. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose. Mr. Rivers and Dr. Powell-Wiley are joint first authors.
- Abbreviations and Acronyms
- body mass index
- 18F-FDG PET/CT
- 18F-Fluorodeoxyglucose positron-emission tomography/computed tomography
- homeostasis model assessment of insulin resistance
- Received June 9, 2017.
- Revision received August 15, 2017.
- Accepted August 16, 2017.
- Eckel R.H.,
- Krauss R.M.
- Rao G.,
- Powell-Wiley T.M.,
- Ancheta I.,
- et al.
- Naik H.B.,
- Natarajan B.,
- Stansky E.,
- et al.
- Dey A.K.,
- Joshi A.A.,
- Chaturvedi A.,
- et al.
- Figueroa A.L.,
- Takx R.A.,
- MacNabb M.H.,
- et al.
- Fleming P.,
- Kraft J.,
- Gulliver W.P.,
- Lynde C.
- Staniak H.L.,
- Bittencourt M.S.,
- de Souza Santos I.,
- et al.
- Rose S.,
- Stansky E.,
- Dagur P.K.,
- et al.
- Li W.G.,
- Gavrila D.,
- Liu X.,
- et al.
- Ouchi N.,
- Kihara S.,
- Arita Y.,
- et al.
- Shuster A.,
- Patlas M.,
- Pinthus J.H.,
- Mourtzakis M.
- Hjuler K.F.,
- Bottcher M.,
- Vestergaard C.,
- Botker H.E.,
- Iversen L.,
- Kragballe K.
- Figueroa A.L.,
- Abdelbaky A.,
- Truong Q.A.,
- et al.
- Fontana L.,
- Eagon J.C.,
- Trujillo M.E.,
- Scherer P.E.,
- Klein S.